initial commit

INSTALL.md

@@ -0,0 +1,46 @@
+### Set up the python environment
+
+```
+conda create -n neuralbody python=3.7
+conda activate neuralbody
+
+# make sure that the pytorch cuda is consistent with the system cuda
+# e.g., if your system cuda is 10.0, install torch 1.4 built from cuda 10.0
+pip install torch==1.4.0+cu100 -f https://download.pytorch.org/whl/torch_stable.html
+
+pip install -r requirements.txt
+
+# install spconv
+cd
+git clone https://github.com/traveller59/spconv --recursive
+cd spconv
+git checkout abf0acf30f5526ea93e687e3f424f62d9cd8313a
+git submodule update --init --recursive
+export CUDA_HOME="/usr/local/cuda-10.0"
+python setup.py bdist_wheel
+cd dist
+pip install spconv-1.2.1-cp36-cp36m-linux_x86_64.whl
+```
+
+### Set up datasets
+
+#### People-Snapshot dataset
+
+1. Download the People-Snapshot dataset [here](https://graphics.tu-bs.de/people-snapshot).
+2. Process the People-Snapshot dataset using the [script](https://github.com/zju3dv/neuralbody#process-people-snapshot).
+3. Create a soft link:
+    ```
+    ROOT=/path/to/neuralbody
+    cd $ROOT/data
+    ln -s /path/to/people_snapshot people_snapshot
+    ```
+
+#### ZJU-Mocap dataset
+
+1. If someone wants to download the ZJU-Mocap dataset, please fill in the [agreement](https://zjueducn-my.sharepoint.com/:b:/g/personal/pengsida_zju_edu_cn/EUPiybrcFeNEhdQROx4-LNEBm4lzLxDwkk1SBcNWFgeplA?e=BGDiQh), and email me (pengsida@zju.edu.cn) and cc Xiaowei Zhou (xwzhou@zju.edu.cn) to request the download link.
+2. Create a soft link:
+    ```
+    ROOT=/path/to/neuralbody
+    cd $ROOT/data
+    ln -s /path/to/zju_mocap zju_mocap
+    ```

LICENSE

@@ -0,0 +1,18 @@
+//////////////////////////////////////////////////////////////////////////// 
+// Copyright 2020-2021 the 3D Vision Group at the State Key Lab of CAD&CG,  
+// Zhejiang University. All Rights Reserved. 
+// 
+// For more information see <https://github.com/zju3dv/neuralbody> 
+// If you use this code, please cite the corresponding publications as  
+// listed on the above website. 
+// 
+// Permission to use, copy, modify and distribute this software and its 
+// documentation for educational, research and non-profit purposes only. 
+// Any modification based on this work must be open source and prohibited 
+// for commercial use. 
+// You must retain, in the source form of any derivative works that you  
+// distribute, all copyright, patent, trademark, and attribution notices  
+// from the source form of this work. 
+//  
+// 
+////////////////////////////////////////////////////////////////////////////

README.md

@@ -1,12 +1,202 @@
----
-title: NeuralBody
-emoji: 📚
-colorFrom: yellow
-colorTo: indigo
-sdk: gradio
-sdk_version: 3.0.10
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces#reference
+**News**
+
+* `05/17/2021` To make the comparison on ZJU-MoCap easier, we save quantitative and qualitative results of other methods at [here](https://github.com/zju3dv/neuralbody/blob/master/supplementary_material.md#results-of-other-methods-on-zju-mocap), including Neural Volumes, Multi-view Neural Human Rendering, and Deferred Neural Human Rendering.
+* `05/13/2021` To make the following works easier compare with our model, we save our rendering results of ZJU-MoCap at [here](https://zjueducn-my.sharepoint.com/:u:/g/personal/pengsida_zju_edu_cn/Ea3VOUy204VAiVJ-V-OGd9YBxdhbtfpS-U6icD_rDq0mUQ?e=cAcylK) and write a [document](supplementary_material.md) that describes the training and test protocols.
+* `05/12/2021` The code supports the test and visualization on unseen human poses.
+* `05/12/2021` We update the ZJU-MoCap dataset with better fitted SMPL using [EasyMocap](https://github.com/zju3dv/EasyMocap). We also release a [website](https://zju3dv.github.io/zju_mocap/) for visualization. Please see [here](https://github.com/zju3dv/neuralbody#potential-problems-of-provided-smpl-parameters) for the usage of provided smpl parameters.
+
+# Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans
+### [Project Page](https://zju3dv.github.io/neuralbody) | [Video](https://www.youtube.com/watch?v=BPCAMeBCE-8) | [Paper](https://arxiv.org/pdf/2012.15838.pdf) | [Data](https://github.com/zju3dv/neuralbody/blob/master/INSTALL.md#zju-mocap-dataset)
+
+![monocular](https://zju3dv.github.io/neuralbody/images/monocular.gif)
+
+> [Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans](https://arxiv.org/pdf/2012.15838.pdf)  
+> Sida Peng, Yuanqing Zhang, Yinghao Xu, Qianqian Wang, Qing Shuai, Hujun Bao, Xiaowei Zhou  
+> CVPR 2021
+
+Any questions or discussions are welcomed!
+
+## Installation
+
+Please see [INSTALL.md](INSTALL.md) for manual installation.
+
+### Installation using docker
+
+Please see [docker/README.md](docker/README.md).
+
+Thanks to [Zhaoyi Wan](https://github.com/wanzysky) for providing the docker implementation.
+
+## Run the code on the custom dataset
+
+Please see [CUSTOM](tools/custom).
+
+## Run the code on People-Snapshot
+
+Please see [INSTALL.md](INSTALL.md) to download the dataset.
+
+We provide the pretrained models at [here](https://zjueducn-my.sharepoint.com/:f:/g/personal/pengsida_zju_edu_cn/Enn43YWDHwBEg-XBqnetFYcBLr3cItZ0qUFU-oKUpDHKXw?e=FObjE9).
+
+### Process People-Snapshot
+
+We already provide some processed data. If you want to process more videos of People-Snapshot, you could use [tools/process_snapshot.py](tools/process_snapshot.py).
+
+You can also visualize smpl parameters of People-Snapshot with [tools/vis_snapshot.py](tools/vis_snapshot.py).
+
+### Visualization on People-Snapshot
+
+Take the visualization on `female-3-casual` as an example. The command lines for visualization are recorded in [visualize.sh](visualize.sh).
+
+1. Download the corresponding pretrained model and put it to `$ROOT/data/trained_model/if_nerf/female3c/latest.pth`.
+2. Visualization:
+    * Visualize novel views of single frame
+    ```
+    python run.py --type visualize --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c vis_novel_view True num_render_views 144
+    ```
+
+    ![monocular](https://zju3dv.github.io/neuralbody/images/monocular_render.gif)
+
+    * Visualize views of dynamic humans with fixed camera
+    ```
+    python run.py --type visualize --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c vis_novel_pose True
+    ```
+
+    ![monocular](https://zju3dv.github.io/neuralbody/images/monocular_perform.gif)
+
+    * Visualize mesh
+    ```
+    # generate meshes
+    python run.py --type visualize --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c vis_mesh True train.num_workers 0
+    # visualize a specific mesh
+    python tools/render_mesh.py --exp_name female3c --dataset people_snapshot --mesh_ind 226
+    ```
+
+    ![monocular](https://zju3dv.github.io/neuralbody/images/monocular_mesh.gif)
+
+3. The results of visualization are located at `$ROOT/data/render/female3c` and `$ROOT/data/perform/female3c`.
+
+### Training on People-Snapshot
+
+Take the training on `female-3-casual` as an example. The command lines for training are recorded in [train.sh](train.sh).
+
+1. Train:
+    ```
+    # training
+    python train_net.py --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c resume False
+    # distributed training
+    python -m torch.distributed.launch --nproc_per_node=4 train_net.py --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c resume False gpus "0, 1, 2, 3" distributed True
+    ```
+2. Train with white background:
+    ```
+    # training
+    python train_net.py --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c resume False white_bkgd True
+    ```
+3. Tensorboard:
+    ```
+    tensorboard --logdir data/record/if_nerf
+    ```
+
+## Run the code on ZJU-MoCap
+
+Please see [INSTALL.md](INSTALL.md) to download the dataset.
+
+We provide the pretrained models at [here](https://zjueducn-my.sharepoint.com/:f:/g/personal/pengsida_zju_edu_cn/Enn43YWDHwBEg-XBqnetFYcBLr3cItZ0qUFU-oKUpDHKXw?e=FObjE9).
+
+### Potential problems of provided smpl parameters
+
+1. The newly fitted parameters locate in `new_params`. Currently, the released pretrained models are trained on previously fitted parameters, which locate in `params`.
+2. The smpl parameters of ZJU-MoCap have different definition from the one of MPI's smplx.
+    * If you want to extract vertices from the provided smpl parameters, please use `zju_smpl/extract_vertices.py`.
+    * The reason that we use the current definition is described at [here](https://github.com/zju3dv/EasyMocap/blob/master/doc/02_output.md#attention-for-smplsmpl-x-users).
+
+It is okay to train Neural Body with smpl parameters fitted by smplx.
+
+### Test on ZJU-MoCap
+
+The command lines for test are recorded in [test.sh](test.sh).
+
+Take the test on `sequence 313` as an example.
+
+1. Download the corresponding pretrained model and put it to `$ROOT/data/trained_model/if_nerf/xyzc_313/latest.pth`.
+2. Test on training human poses:
+    ```
+    python run.py --type evaluate --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313
+    ```
+3. Test on unseen human poses:
+    ```
+    python run.py --type evaluate --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 test_novel_pose True
+    ```
+
+### Visualization on ZJU-MoCap
+
+Take the visualization on `sequence 313` as an example. The command lines for visualization are recorded in [visualize.sh](visualize.sh).
+
+1. Download the corresponding pretrained model and put it to `$ROOT/data/trained_model/if_nerf/xyzc_313/latest.pth`.
+2. Visualization:
+    * Visualize novel views of single frame
+    ```
+    python run.py --type visualize --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 vis_novel_view True
+    ```
+    ![zju_mocap](https://zju3dv.github.io/neuralbody/images/zju_mocap_render_313.gif)
+
+    * Visualize novel views of single frame by rotating the SMPL model
+    ```
+    python run.py --type visualize --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 vis_novel_view True num_render_views 100
+    ```
+    ![zju_mocap](https://zju3dv.github.io/neuralbody/images/rotate_smpl.gif)
+
+    * Visualize views of dynamic humans with fixed camera
+    ```
+    python run.py --type visualize --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 vis_novel_pose True num_render_frame 1000 num_render_views 1
+    ```
+    ![zju_mocap](https://zju3dv.github.io/neuralbody/images/zju_mocap_perform_fixed_313.gif) 
+
+    * Visualize views of dynamic humans with rotated camera
+    ```
+    python run.py --type visualize --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 vis_novel_pose True num_render_frame 1000
+    ```
+    ![zju_mocap](https://zju3dv.github.io/neuralbody/images/zju_mocap_perform_313.gif)
+
+    * Visualize mesh
+    ```
+    # generate meshes
+    python run.py --type visualize --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 vis_mesh True train.num_workers 0
+    # visualize a specific mesh
+    python tools/render_mesh.py --exp_name xyzc_313 --dataset zju_mocap --mesh_ind 0
+    ```
+    ![zju_mocap](https://zju3dv.github.io/neuralbody/images/zju_mocap_mesh.gif)
+
+4. The results of visualization are located at `$ROOT/data/render/xyzc_313` and `$ROOT/data/perform/xyzc_313`.
+
+### Training on ZJU-MoCap
+
+Take the training on `sequence 313` as an example. The command lines for training are recorded in [train.sh](train.sh).
+
+1. Train:
+    ```
+    # training
+    python train_net.py --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 resume False
+    # distributed training
+    python -m torch.distributed.launch --nproc_per_node=4 train_net.py --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 resume False gpus "0, 1, 2, 3" distributed True
+    ```
+2. Train with white background:
+    ```
+    # training
+    python train_net.py --cfg_file configs/zju_mocap_exp/latent_xyzc_313.yaml exp_name xyzc_313 resume False white_bkgd True
+    ```
+3. Tensorboard:
+    ```
+    tensorboard --logdir data/record/if_nerf
+    ```
+
+## Citation
+
+If you find this code useful for your research, please use the following BibTeX entry.
+
+```
+@inproceedings{peng2021neural,
+  title={Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans},
+  author={Peng, Sida and Zhang, Yuanqing and Xu, Yinghao and Wang, Qianqian and Shuai, Qing and Bao, Hujun and Zhou, Xiaowei},
+  booktitle={CVPR},
+  year={2021}
+}
+```

configs/h36m_exp/latent_xyzc_s11g.yaml

@@ -0,0 +1,28 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+train_dataset:
+    data_root: 'data/h36m/S11/Greeting'
+    human: 'S11'
+    ann_file: 'data/h36m/S11/Greeting/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/h36m/S11/Greeting'
+    human: 'S11'
+    ann_file: 'data/h36m/S11/Greeting/annots.npy'
+    split: 'test'
+
+# data options
+H: 1002
+W: 1000
+ratio: 1.
+training_view: [0, 1, 2, 3]
+begin_ith_frame: 1200
+num_train_frame: 400
+smpl: 'smpl'
+vertices: 'vertices'
+params: 'params'
+big_box: True

configs/h36m_exp/latent_xyzc_s9p.yaml

@@ -0,0 +1,28 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+train_dataset:
+    data_root: 'data/h36m/S9/Posing'
+    human: 'S9'
+    ann_file: 'data/h36m/S9/Posing/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/h36m/S9/Posing'
+    human: 'S9'
+    ann_file: 'data/h36m/S9/Posing/annots.npy'
+    split: 'test'
+
+# data options
+H: 1002
+W: 1000
+ratio: 1.
+training_view: [0, 1, 2, 3]
+begin_ith_frame: 1000
+num_train_frame: 300
+smpl: 'smpl'
+vertices: 'vertices'
+params: 'params'
+big_box: True

configs/monocular_custom.yaml

@@ -0,0 +1,25 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'path/to/custom_data',
+    human: 'custom',
+    ann_file: 'path/to/custom_data/params.npy',
+    split: 'train'
+
+test_dataset:
+    data_root: 'path/to/custom_data',
+    human: 'custom',
+    ann_file: 'path/to/custom_data/params.npy',
+    split: 'test'
+
+# data options
+ratio: 1.
+training_view: [0, 6, 12, 18]
+num_train_frame: 300
+smpl: 'smpl'
+vertices: 'vertices'
+params: 'params'
+big_box: True

configs/multi_view_custom.yaml

@@ -0,0 +1,25 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+train_dataset:
+    data_root: 'path/to/custom_data',
+    human: 'custom',
+    ann_file: 'path/to/custom_data/annots.npy',
+    split: 'train'
+
+test_dataset:
+    data_root: 'path/to/custom_data',
+    human: 'custom',
+    ann_file: 'path/to/custom_data/annots.npy',
+    split: 'test'
+
+# data options
+ratio: 1.
+training_view: [0, 6, 12, 18]
+num_train_frame: 300
+smpl: 'smpl'
+vertices: 'vertices'
+params: 'params'
+big_box: True

configs/nerf/nerf_313.yaml

@@ -0,0 +1,145 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.multi_view_dataset'
+train_dataset_path: 'lib/datasets/light_stage/multi_view_dataset.py'
+test_dataset_module: 'lib.datasets.light_stage.multi_view_dataset'
+test_dataset_path: 'lib/datasets/light_stage/multi_view_dataset.py'
+
+network_module: 'lib.networks.nerf'
+network_path: 'lib/networks/nerf.py'
+renderer_module: 'lib.networks.renderer.volume_renderer'
+renderer_path: 'lib/networks/renderer/volume_renderer.py'
+
+trainer_module: 'lib.train.trainers.nerf.py'
+trainer_path: 'lib/train/trainers/nerf.py'
+
+evaluator_module: 'lib.evaluators.if_nerf'
+evaluator_path: 'lib/evaluators/if_nerf.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 313
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'test'
+
+train:
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+lindisp: False
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+perturb: 1
+white_bkgd: False
+
+num_render_views: 50
+
+# data options
+ratio: 0.5
+num_train_frame: 1
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1
+
+
+novel_view_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_demo_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_demo_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_demo_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_demo_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.volume_renderer'
+    renderer_path: 'lib/networks/renderer/volume_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_demo'
+    visualizer_path: 'lib/visualizers/if_nerf_demo.py'
+
+    test:
+        sampler: ''
+
+novel_pose_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_perform_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_perform_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_perform_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_perform_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.volume_renderer'
+    renderer_path: 'lib/networks/renderer/volume_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_perform'
+    visualizer_path: 'lib/visualizers/if_nerf_perform.py'
+
+    test:
+        sampler: ''
+
+mesh_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_mesh_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_mesh_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_mesh_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_mesh_dataset.py'
+
+    network_module: 'lib.networks.latent_xyzc'
+    network_path: 'lib/networks/latent_xyzc.py'
+    renderer_module: 'lib.networks.renderer.volume_mesh_renderer'
+    renderer_path: 'lib/networks/renderer/volume_mesh_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_mesh'
+    visualizer_path: 'lib/visualizers/if_nerf_mesh.py'
+
+    mesh_th: 5
+
+    test:
+        sampler: 'FrameSampler'
+        frame_sampler_interval: 1

configs/nerf/nerf_315.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 315
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'test'

configs/nerf/nerf_377.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 377
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_377'
+    human: 'CoreView_377'
+    ann_file: 'data/zju_mocap/CoreView_377/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_377'
+    human: 'CoreView_377'
+    ann_file: 'data/zju_mocap/CoreView_377/annots.npy'
+    split: 'test'

configs/nerf/nerf_386.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 386
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_386'
+    human: 'CoreView_386'
+    ann_file: 'data/zju_mocap/CoreView_386/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_386'
+    human: 'CoreView_386'
+    ann_file: 'data/zju_mocap/CoreView_386/annots.npy'
+    split: 'test'

configs/nerf/nerf_387.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 387
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_387'
+    human: 'CoreView_387'
+    ann_file: 'data/zju_mocap/CoreView_387/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_387'
+    human: 'CoreView_387'
+    ann_file: 'data/zju_mocap/CoreView_387/annots.npy'
+    split: 'test'

configs/nerf/nerf_390.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 390
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_390'
+    human: 'CoreView_390'
+    ann_file: 'data/zju_mocap/CoreView_390/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_390'
+    human: 'CoreView_390'
+    ann_file: 'data/zju_mocap/CoreView_390/annots.npy'
+    split: 'test'

configs/nerf/nerf_392.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 392
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_392'
+    human: 'CoreView_392'
+    ann_file: 'data/zju_mocap/CoreView_392/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_392'
+    human: 'CoreView_392'
+    ann_file: 'data/zju_mocap/CoreView_392/annots.npy'
+    split: 'test'

configs/nerf/nerf_393.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 393
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_393'
+    human: 'CoreView_393'
+    ann_file: 'data/zju_mocap/CoreView_393/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_393'
+    human: 'CoreView_393'
+    ann_file: 'data/zju_mocap/CoreView_393/annots.npy'
+    split: 'test'

configs/nerf/nerf_394.yaml

@@ -0,0 +1,18 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/nerf/nerf_313.yaml'
+
+human: 394
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_394'
+    human: 'CoreView_394'
+    ann_file: 'data/zju_mocap/CoreView_394/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_394'
+    human: 'CoreView_394'
+    ann_file: 'data/zju_mocap/CoreView_394/annots.npy'
+    split: 'test'

configs/neural_volumes/neural_volumes_313.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 313
+
+train:
+    dataset: Human313_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human313_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 60
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_315.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 315
+
+train:
+    dataset: Human315_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human315_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 400
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_377.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 377
+
+train:
+    dataset: Human377_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human377_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_386.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 386
+
+train:
+    dataset: Human386_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human386_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_387.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 387
+
+train:
+    dataset: Human387_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human387_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_390.yaml

@@ -0,0 +1,95 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 390
+
+train:
+    dataset: Human390_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human390_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+begin_i: 700
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_392.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 392
+
+train:
+    dataset: Human392_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human392_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_393.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 393
+
+train:
+    dataset: Human393_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human393_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/neural_volumes/neural_volumes_394.yaml

@@ -0,0 +1,94 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+test_dataset_module: 'lib.datasets.light_stage.can_smpl'
+test_dataset_path: 'lib/datasets/light_stage/can_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.neural_volume'
+evaluator_path: 'lib/evaluators/neural_volume.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 394
+
+train:
+    dataset: Human394_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human394_0001_Test
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 300
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/snapshot_exp/snapshot_f1c.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-1-casual'
+    human: 'female-1-casual'
+    ann_file: 'data/people_snapshot/female-1-casual/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-1-casual'
+    human: 'female-1-casual'
+    ann_file: 'data/people_snapshot/female-1-casual/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 250

configs/snapshot_exp/snapshot_f3c.yaml

@@ -0,0 +1,134 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.monocular_dataset'
+train_dataset_path: 'lib/datasets/light_stage/monocular_dataset.py'
+test_dataset_module: 'lib.datasets.light_stage.monocular_dataset'
+test_dataset_path: 'lib/datasets/light_stage/monocular_dataset.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.if_nerf'
+evaluator_path: 'lib/evaluators/if_nerf.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-3-casual'
+    human: 'female-3-casual'
+    ann_file: 'data/people_snapshot/female-3-casual/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-3-casual'
+    human: 'female-3-casual'
+    ann_file: 'data/people_snapshot/female-3-casual/params.npy'
+    split: 'test'
+
+train:
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 100
+eval_ep: 1000
+
+# rendering options
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+perturb: 1
+white_bkgd: False
+
+num_render_views: 50
+
+# data options
+H: 1080
+W: 1080
+ratio: 1.
+num_train_frame: 230
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1
+
+
+novel_view_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.monocular_demo_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/monocular_demo_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.monocular_demo_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/monocular_demo_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.if_clight_renderer_msk'
+    renderer_path: 'lib/networks/renderer/if_clight_renderer_msk.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_demo'
+    visualizer_path: 'lib/visualizers/if_nerf_demo.py'
+
+    ratio: 0.5
+
+    test:
+        sampler: ''
+
+novel_pose_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.monocular_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/monocular_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.monocular_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/monocular_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.if_clight_renderer_msk'
+    renderer_path: 'lib/networks/renderer/if_clight_renderer_msk.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_perform'
+    visualizer_path: 'lib/visualizers/if_nerf_perform.py'
+
+    ratio: 0.5
+
+    test:
+        sampler: ''
+
+mesh_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.monocular_mesh_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/monocular_mesh_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.monocular_mesh_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/monocular_mesh_dataset.py'
+
+    network_module: 'lib.networks.latent_xyzc'
+    network_path: 'lib/networks/latent_xyzc.py'
+    renderer_module: 'lib.networks.renderer.if_mesh_renderer'
+    renderer_path: 'lib/networks/renderer/if_mesh_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_mesh'
+    visualizer_path: 'lib/visualizers/if_nerf_mesh.py'
+
+    mesh_th: 5
+
+    test:
+        sampler: 'FrameSampler'
+        frame_sampler_interval: 1

configs/snapshot_exp/snapshot_f4c.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-4-casual'
+    human: 'female-4-casual'
+    ann_file: 'data/people_snapshot/female-4-casual/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-4-casual'
+    human: 'female-4-casual'
+    ann_file: 'data/people_snapshot/female-4-casual/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 200
+begin_ith_frame: 10

configs/snapshot_exp/snapshot_f6p.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-6-plaza'
+    human: 'female-6-plaza'
+    ann_file: 'data/people_snapshot/female-6-plaza/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-6-plaza'
+    human: 'female-6-plaza'
+    ann_file: 'data/people_snapshot/female-6-plaza/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 240

configs/snapshot_exp/snapshot_f7p.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-7-plaza'
+    human: 'female-7-plaza'
+    ann_file: 'data/people_snapshot/female-7-plaza/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-7-plaza'
+    human: 'female-7-plaza'
+    ann_file: 'data/people_snapshot/female-7-plaza/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 185

configs/snapshot_exp/snapshot_f8p.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/female-8-plaza'
+    human: 'female-8-plaza'
+    ann_file: 'data/people_snapshot/female-8-plaza/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/female-8-plaza'
+    human: 'female-8-plaza'
+    ann_file: 'data/people_snapshot/female-8-plaza/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 200

configs/snapshot_exp/snapshot_m2c.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/male-2-casual'
+    human: 'male-2-casual'
+    ann_file: 'data/people_snapshot/male-2-casual/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/male-2-casual'
+    human: 'male-2-casual'
+    ann_file: 'data/people_snapshot/male-2-casual/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 180

configs/snapshot_exp/snapshot_m2o.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/male-2-outdoor'
+    human: 'male-2-outdoor'
+    ann_file: 'data/people_snapshot/male-2-outdoor/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/male-2-outdoor'
+    human: 'male-2-outdoor'
+    ann_file: 'data/people_snapshot/male-2-outdoor/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 150

configs/snapshot_exp/snapshot_m3c.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/male-3-casual'
+    human: 'male-3-casual'
+    ann_file: 'data/people_snapshot/male-3-casual/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/male-3-casual'
+    human: 'male-3-casual'
+    ann_file: 'data/people_snapshot/male-3-casual/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 235

configs/snapshot_exp/snapshot_m5o.yaml

@@ -0,0 +1,20 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/snapshot_exp/snapshot_f3c.yaml'
+
+train_dataset:
+    data_root: 'data/people_snapshot/male-5-outdoor'
+    human: 'male-5-outdoor'
+    ann_file: 'data/people_snapshot/male-5-outdoor/params.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/people_snapshot/male-5-outdoor'
+    human: 'male-5-outdoor'
+    ann_file: 'data/people_snapshot/male-5-outdoor/params.npy'
+    split: 'test'
+
+# data options
+ratio: 1.
+num_train_frame: 295

configs/zju_mocap_exp/latent_xyzc_313.yaml

@@ -0,0 +1,152 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.multi_view_dataset'
+train_dataset_path: 'lib/datasets/light_stage/multi_view_dataset.py'
+test_dataset_module: 'lib.datasets.light_stage.multi_view_dataset'
+test_dataset_path: 'lib/datasets/light_stage/multi_view_dataset.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.if_nerf'
+evaluator_path: 'lib/evaluators/if_nerf.py'
+
+visualizer_module: 'lib.visualizers.if_nerf'
+visualizer_path: 'lib/visualizers/if_nerf.py'
+
+human: 313
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'test'
+
+train:
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    sampler: 'FrameSampler'
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# rendering options
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+perturb: 1
+white_bkgd: False
+
+num_render_views: 50
+
+# data options
+H: 1024
+W: 1024
+ratio: 0.5
+training_view: [0, 6, 12, 18]
+num_train_frame: 60
+num_novel_pose_frame: 1000
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1
+
+
+novel_view_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_demo_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_demo_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_demo_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_demo_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.if_clight_renderer_mmsk'
+    renderer_path: 'lib/networks/renderer/if_clight_renderer_mmsk.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_demo'
+    visualizer_path: 'lib/visualizers/if_nerf_demo.py'
+
+    test:
+        sampler: ''
+
+rotate_smpl_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.rotate_smpl_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/rotate_smpl_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.rotate_smpl_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/rotate_smpl_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.if_clight_renderer'
+    renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_demo'
+    visualizer_path: 'lib/visualizers/if_nerf_demo.py'
+
+    test:
+        sampler: ''
+
+novel_pose_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_perform_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_perform_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_perform_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_perform_dataset.py'
+
+    renderer_module: 'lib.networks.renderer.if_clight_renderer_mmsk'
+    renderer_path: 'lib/networks/renderer/if_clight_renderer_mmsk.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_perform'
+    visualizer_path: 'lib/visualizers/if_nerf_perform.py'
+
+    test:
+        sampler: ''
+
+mesh_cfg:
+    train_dataset_module: 'lib.datasets.light_stage.multi_view_mesh_dataset'
+    train_dataset_path: 'lib/datasets/light_stage/multi_view_mesh_dataset.py'
+    test_dataset_module: 'lib.datasets.light_stage.multi_view_mesh_dataset'
+    test_dataset_path: 'lib/datasets/light_stage/multi_view_mesh_dataset.py'
+
+    network_module: 'lib.networks.latent_xyzc'
+    network_path: 'lib/networks/latent_xyzc.py'
+    renderer_module: 'lib.networks.renderer.if_mesh_renderer'
+    renderer_path: 'lib/networks/renderer/if_mesh_renderer.py'
+
+    visualizer_module: 'lib.visualizers.if_nerf_mesh'
+    visualizer_path: 'lib/visualizers/if_nerf_mesh.py'
+
+    mesh_th: 5
+
+    test:
+        sampler: 'FrameSampler'
+        frame_sampler_interval: 1

configs/zju_mocap_exp/latent_xyzc_315.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 315
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 400

configs/zju_mocap_exp/latent_xyzc_377.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 377
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_377'
+    human: 'CoreView_377'
+    ann_file: 'data/zju_mocap/CoreView_377/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_377'
+    human: 'CoreView_377'
+    ann_file: 'data/zju_mocap/CoreView_377/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_386.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 386
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_386'
+    human: 'CoreView_386'
+    ann_file: 'data/zju_mocap/CoreView_386/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_386'
+    human: 'CoreView_386'
+    ann_file: 'data/zju_mocap/CoreView_386/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_387.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 387
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_387'
+    human: 'CoreView_387'
+    ann_file: 'data/zju_mocap/CoreView_387/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_387'
+    human: 'CoreView_387'
+    ann_file: 'data/zju_mocap/CoreView_387/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_390.yaml

@@ -0,0 +1,23 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 390
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_390'
+    human: 'CoreView_390'
+    ann_file: 'data/zju_mocap/CoreView_390/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_390'
+    human: 'CoreView_390'
+    ann_file: 'data/zju_mocap/CoreView_390/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300
+begin_ith_frame: 700
+num_novel_pose_frame: 700

configs/zju_mocap_exp/latent_xyzc_392.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 392
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_392'
+    human: 'CoreView_392'
+    ann_file: 'data/zju_mocap/CoreView_392/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_392'
+    human: 'CoreView_392'
+    ann_file: 'data/zju_mocap/CoreView_392/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_393.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 393
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_393'
+    human: 'CoreView_393'
+    ann_file: 'data/zju_mocap/CoreView_393/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_393'
+    human: 'CoreView_393'
+    ann_file: 'data/zju_mocap/CoreView_393/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_394.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 394
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_394'
+    human: 'CoreView_394'
+    ann_file: 'data/zju_mocap/CoreView_394/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_394'
+    human: 'CoreView_394'
+    ann_file: 'data/zju_mocap/CoreView_394/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_395.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 395
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_395'
+    human: 'CoreView_395'
+    ann_file: 'data/zju_mocap/CoreView_395/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_395'
+    human: 'CoreView_395'
+    ann_file: 'data/zju_mocap/CoreView_395/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 300

configs/zju_mocap_exp/latent_xyzc_396.yaml

@@ -0,0 +1,22 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 396
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_396'
+    human: 'CoreView_396'
+    ann_file: 'data/zju_mocap/CoreView_396/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_396'
+    human: 'CoreView_396'
+    ann_file: 'data/zju_mocap/CoreView_396/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 540
+begin_ith_frame: 810

configs/zju_mocap_exp/xyzc_rotate_demo_313.yaml

@@ -0,0 +1,93 @@
+task: 'if_nerf'
+gpus: [0]
+
+train_dataset_module: 'lib.datasets.light_stage.can_smpl_demo'
+train_dataset_path: 'lib/datasets/light_stage/can_smpl_demo.py'
+test_dataset_module: 'lib.datasets.light_stage.rotate_smpl'
+test_dataset_path: 'lib/datasets/light_stage/rotate_smpl.py'
+
+network_module: 'lib.networks.latent_xyzc'
+network_path: 'lib/networks/latent_xyzc.py'
+renderer_module: 'lib.networks.renderer.if_clight_renderer'
+renderer_path: 'lib/networks/renderer/if_clight_renderer.py'
+
+trainer_module: 'lib.train.trainers.if_nerf_clight'
+trainer_path: 'lib/train/trainers/if_nerf_clight.py'
+
+evaluator_module: 'lib.evaluators.if_nerf'
+evaluator_path: 'lib/evaluators/if_nerf.py'
+
+visualizer_module: 'lib.visualizers.if_nerf_demo'
+visualizer_path: 'lib/visualizers/if_nerf_demo.py'
+
+human: 313
+
+train:
+    dataset: Human313_0001_Train
+    batch_size: 1
+    collator: ''
+    lr: 5e-4
+    weight_decay: 0
+    epoch: 400
+    scheduler:
+        type: 'exponential'
+        gamma: 0.1
+        decay_epochs: 1000
+    num_workers: 16
+
+test:
+    dataset: Human313_0001_Test
+    batch_size: 1
+    collator: ''
+
+ep_iter: 500
+save_ep: 1000
+eval_ep: 1000
+
+# training options
+netdepth: 8
+netwidth: 256
+netdepth_fine: 8
+netwidth_fine: 256
+netchunk: 65536
+chunk: 32768
+
+no_batching: True
+
+precrop_iters: 500
+precrop_frac: 0.5
+
+# network options
+point_feature: 6
+
+# rendering options
+use_viewdirs: True
+i_embed: 0
+xyz_res: 10
+view_res: 4
+raw_noise_std: 0
+
+N_samples: 64
+N_importance: 128
+N_rand: 1024
+
+near: 1
+far: 3
+
+perturb: 1
+white_bkgd: False
+
+render_views: 50
+
+# data options
+res: 256
+ratio: 0.5
+intv: 6
+ni: 60
+smpl: 'smpl'
+params: 'params'
+
+voxel_size: [0.005, 0.005, 0.005]  # dhw
+
+# record options
+log_interval: 1

configs/zju_mocap_frame1_exp/latent_xyzc_313_ni1.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 313
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_313'
+    human: 'CoreView_313'
+    ann_file: 'data/zju_mocap/CoreView_313/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 1

configs/zju_mocap_frame1_exp/latent_xyzc_315_ni1.yaml

@@ -0,0 +1,21 @@
+task: 'if_nerf'
+gpus: [0]
+
+parent_cfg: 'configs/zju_mocap_exp/latent_xyzc_313.yaml'
+
+human: 315
+
+train_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'train'
+
+test_dataset:
+    data_root: 'data/zju_mocap/CoreView_315'
+    human: 'CoreView_315'
+    ann_file: 'data/zju_mocap/CoreView_315/annots.npy'
+    split: 'test'
+
+# data options
+num_train_frame: 1