Upload 5 files

README.md

@@ -1,13 +1,46 @@
 ---
-title: S19erav1
-emoji: 📊
-colorFrom: green
-colorTo: red
+title: FastSAM
+emoji: 🐠
+colorFrom: pink
+colorTo: indigo
 sdk: gradio
-sdk_version: 3.46.1
-app_file: app.py
+sdk_version: 3.45.1
+app_file: app_gradio.py
 pinned: false
-license: mit
+license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Fast Segment Anything
+
+Official PyTorch Implementation of the <a href="https://github.com/CASIA-IVA-Lab/FastSAM">.
+
+The **Fast Segment Anything Model(FastSAM)** is a CNN Segment Anything Model trained by only 2% of the SA-1B dataset published by SAM authors. The FastSAM achieve a comparable performance with
+the SAM method at **50× higher run-time speed**.
+
+
+## License
+
+The model is licensed under the [Apache 2.0 license](LICENSE).
+
+
+## Acknowledgement
+
+- [Segment Anything](https://segment-anything.com/) provides the SA-1B dataset and the base codes.
+- [YOLOv8](https://github.com/ultralytics/ultralytics) provides codes and pre-trained models.
+- [YOLACT](https://arxiv.org/abs/2112.10003) provides powerful instance segmentation method.
+- [Grounded-Segment-Anything](https://huggingface.co/spaces/yizhangliu/Grounded-Segment-Anything) provides a useful web demo template.
+
+## Citing FastSAM
+
+If you find this project useful for your research, please consider citing the following BibTeX entry.
+
+```
+@misc{zhao2023fast,
+      title={Fast Segment Anything}, 
+      author={Xu Zhao and Wenchao Ding and Yongqi An and Yinglong Du and Tao Yu and Min Li and Ming Tang and Jinqiao Wang},
+      year={2023},
+      eprint={2306.12156},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV}
+}
+```

app_gradio.py

@@ -0,0 +1,374 @@
+from ultralytics import YOLO
+import gradio as gr
+import torch
+from utils.tools_gradio import fast_process
+from utils.tools import format_results, box_prompt, point_prompt, text_prompt
+from PIL import ImageDraw
+import numpy as np
+
+# Load the pre-trained model
+model = YOLO('./weights/FastSAM.pt')
+
+device = torch.device(
+    "cuda"
+    if torch.cuda.is_available()
+    else "mps"
+    if torch.backends.mps.is_available()
+    else "cpu"
+)
+
+# Description
+title = "<center><strong><font size='8'>🏃 Fast Segment Anything 🤗</font></strong></center>"
+
+news = """ # 📖 News
+        🔥 2023/07/14: Add a "wider result" button in text mode (Thanks for [gaoxinge](https://github.com/CASIA-IVA-Lab/FastSAM/pull/95)).
+
+        🔥 2023/06/29: Support the text mode (Thanks for [gaoxinge](https://github.com/CASIA-IVA-Lab/FastSAM/pull/47)).
+
+        🔥 2023/06/26: Support the points mode. (Better and faster interaction will come soon!)
+
+        🔥 2023/06/24: Add the 'Advanced options" in Everything mode to get a more detailed adjustment.        
+        """  
+
+description_e = """This is a demo on Github project 🏃 [Fast Segment Anything Model](https://github.com/CASIA-IVA-Lab/FastSAM). Welcome to give a star ⭐️ to it.
+                
+                🎯 Upload an Image, segment it with Fast Segment Anything (Everything mode). The other modes will come soon.
+                
+                ⌛️ It takes about 6~ seconds to generate segment results. The concurrency_count of queue is 1, please wait for a moment when it is crowded.
+                
+                🚀 To get faster results, you can use a smaller input size and leave high_visual_quality unchecked.
+                
+                📣 You can also obtain the segmentation results of any Image through this Colab: [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1oX14f6IneGGw612WgVlAiy91UHwFAvr9?usp=sharing)
+                
+                😚 A huge thanks goes out to the @HuggingFace Team for supporting us with GPU grant.
+                
+                🏠 Check out our [Model Card 🏃](https://huggingface.co/An-619/FastSAM)
+                
+              """
+
+description_p = """ # 🎯 Instructions for points mode
+                This is a demo on Github project 🏃 [Fast Segment Anything Model](https://github.com/CASIA-IVA-Lab/FastSAM). Welcome to give a star ⭐️ to it.
+                
+                1. Upload an image or choose an example.
+                
+                2. Choose the point label ('Add mask' means a positive point. 'Remove' Area means a negative point that is not segmented).
+                
+                3. Add points one by one on the image.
+                
+                4. Click the 'Segment with points prompt' button to get the segmentation results.
+                
+                **5. If you get Error, click the 'Clear points' button and try again may help.**
+                
+              """
+
+examples = [["examples/sa_8776.jpg"], ["examples/sa_414.jpg"], ["examples/sa_1309.jpg"], ["examples/sa_11025.jpg"],
+            ["examples/sa_561.jpg"], ["examples/sa_192.jpg"], ["examples/sa_10039.jpg"], ["examples/sa_862.jpg"]]
+
+default_example = examples[0]
+
+css = "h1 { text-align: center } .about { text-align: justify; padding-left: 10%; padding-right: 10%; }"
+
+
+def segment_everything(
+    input,
+    input_size=1024, 
+    iou_threshold=0.7,
+    conf_threshold=0.25,
+    better_quality=False,
+    withContours=True,
+    use_retina=True,
+    text="",
+    wider=False,
+    mask_random_color=True,
+):
+    input_size = int(input_size)  # 确保 imgsz 是整数
+    # Thanks for the suggestion by hysts in HuggingFace.
+    w, h = input.size
+    scale = input_size / max(w, h)
+    new_w = int(w * scale)
+    new_h = int(h * scale)
+    input = input.resize((new_w, new_h))
+
+    results = model(input,
+                    device=device,
+                    retina_masks=True,
+                    iou=iou_threshold,
+                    conf=conf_threshold,
+                    imgsz=input_size,)
+
+    if len(text) > 0:
+        results = format_results(results[0], 0)
+        annotations, _ = text_prompt(results, text, input, device=device, wider=wider)
+        annotations = np.array([annotations])
+    else:
+        annotations = results[0].masks.data
+    
+    fig = fast_process(annotations=annotations,
+                       image=input,
+                       device=device,
+                       scale=(1024 // input_size),
+                       better_quality=better_quality,
+                       mask_random_color=mask_random_color,
+                       bbox=None,
+                       use_retina=use_retina,
+                       withContours=withContours,)
+    return fig
+
+
+def segment_with_points(
+    input,
+    input_size=1024, 
+    iou_threshold=0.7,
+    conf_threshold=0.25,
+    better_quality=False,
+    withContours=True,
+    use_retina=True,
+    mask_random_color=True,
+):
+    global global_points
+    global global_point_label
+    
+    input_size = int(input_size)  # 确保 imgsz 是整数
+    # Thanks for the suggestion by hysts in HuggingFace.
+    w, h = input.size
+    scale = input_size / max(w, h)
+    new_w = int(w * scale)
+    new_h = int(h * scale)
+    input = input.resize((new_w, new_h))
+    
+    scaled_points = [[int(x * scale) for x in point] for point in global_points]
+
+    results = model(input,
+                    device=device,
+                    retina_masks=True,
+                    iou=iou_threshold,
+                    conf=conf_threshold,
+                    imgsz=input_size,)
+    
+    results = format_results(results[0], 0)
+    annotations, _ = point_prompt(results, scaled_points, global_point_label, new_h, new_w)
+    annotations = np.array([annotations])
+
+    fig = fast_process(annotations=annotations,
+                       image=input,
+                       device=device,
+                       scale=(1024 // input_size),
+                       better_quality=better_quality,
+                       mask_random_color=mask_random_color,
+                       bbox=None,
+                       use_retina=use_retina,
+                       withContours=withContours,)
+
+    global_points = []
+    global_point_label = []
+    return fig, None
+
+
+def get_points_with_draw(image, label, evt: gr.SelectData):
+    global global_points
+    global global_point_label
+
+    x, y = evt.index[0], evt.index[1]
+    point_radius, point_color = 15, (255, 255, 0) if label == 'Add Mask' else (255, 0, 255)
+    global_points.append([x, y])
+    global_point_label.append(1 if label == 'Add Mask' else 0)
+    
+    print(x, y, label == 'Add Mask')
+    
+    # 创建一个可以在图像上绘图的对象
+    draw = ImageDraw.Draw(image)
+    draw.ellipse([(x - point_radius, y - point_radius), (x + point_radius, y + point_radius)], fill=point_color)
+    return image
+
+
+cond_img_e = gr.Image(label="Input", value=default_example[0], type='pil')
+cond_img_p = gr.Image(label="Input with points", value=default_example[0], type='pil')
+cond_img_t = gr.Image(label="Input with text", value="examples/dogs.jpg", type='pil')
+
+segm_img_e = gr.Image(label="Segmented Image", interactive=False, type='pil')
+segm_img_p = gr.Image(label="Segmented Image with points", interactive=False, type='pil')
+segm_img_t = gr.Image(label="Segmented Image with text", interactive=False, type='pil')
+
+global_points = []
+global_point_label = []
+
+input_size_slider = gr.components.Slider(minimum=512,
+                                         maximum=1024,
+                                         value=1024,
+                                         step=64,
+                                         label='Input_size',
+                                         info='Our model was trained on a size of 1024')
+
+with gr.Blocks(css=css, title='Fast Segment Anything') as demo:
+    with gr.Row():
+        with gr.Column(scale=1):
+            # Title
+            gr.Markdown(title)
+
+        with gr.Column(scale=1):
+            # News
+            gr.Markdown(news)
+
+    with gr.Tab("Everything mode"):
+        # Images
+        with gr.Row(variant="panel"):
+            with gr.Column(scale=1):
+                cond_img_e.render()
+
+            with gr.Column(scale=1):
+                segm_img_e.render()
+
+        # Submit & Clear
+        with gr.Row():
+            with gr.Column():
+                input_size_slider.render()
+
+                with gr.Row():
+                    contour_check = gr.Checkbox(value=True, label='withContours', info='draw the edges of the masks')
+
+                    with gr.Column():
+                        segment_btn_e = gr.Button("Segment Everything", variant='primary')
+                        clear_btn_e = gr.Button("Clear", variant="secondary")
+
+                gr.Markdown("Try some of the examples below ⬇️")
+                gr.Examples(examples=examples,
+                            inputs=[cond_img_e],
+                            outputs=segm_img_e,
+                            fn=segment_everything,
+                            cache_examples=True,
+                            examples_per_page=4)
+
+            with gr.Column():
+                with gr.Accordion("Advanced options", open=False):
+                    iou_threshold = gr.Slider(0.1, 0.9, 0.7, step=0.1, label='iou', info='iou threshold for filtering the annotations')
+                    conf_threshold = gr.Slider(0.1, 0.9, 0.25, step=0.05, label='conf', info='object confidence threshold')
+                    with gr.Row():
+                        mor_check = gr.Checkbox(value=False, label='better_visual_quality', info='better quality using morphologyEx')
+                        with gr.Column():
+                            retina_check = gr.Checkbox(value=True, label='use_retina', info='draw high-resolution segmentation masks')
+
+                # Description
+                gr.Markdown(description_e)
+
+    segment_btn_e.click(segment_everything,
+                        inputs=[
+                            cond_img_e,
+                            input_size_slider,
+                            iou_threshold,
+                            conf_threshold,
+                            mor_check,
+                            contour_check,
+                            retina_check,
+                        ],
+                        outputs=segm_img_e)
+
+    with gr.Tab("Points mode"):
+        # Images
+        with gr.Row(variant="panel"):
+            with gr.Column(scale=1):
+                cond_img_p.render()
+
+            with gr.Column(scale=1):
+                segm_img_p.render()
+                
+        # Submit & Clear
+        with gr.Row():
+            with gr.Column():
+                with gr.Row():
+                    add_or_remove = gr.Radio(["Add Mask", "Remove Area"], value="Add Mask", label="Point_label (foreground/background)")
+
+                    with gr.Column():
+                        segment_btn_p = gr.Button("Segment with points prompt", variant='primary')
+                        clear_btn_p = gr.Button("Clear points", variant='secondary')
+
+                gr.Markdown("Try some of the examples below ⬇️")
+                gr.Examples(examples=examples,
+                            inputs=[cond_img_p],
+                            # outputs=segm_img_p,
+                            # fn=segment_with_points,
+                            # cache_examples=True,
+                            examples_per_page=4)
+
+            with gr.Column():
+                # Description
+                gr.Markdown(description_p)
+
+    cond_img_p.select(get_points_with_draw, [cond_img_p, add_or_remove], cond_img_p)
+
+    segment_btn_p.click(segment_with_points,
+                        inputs=[cond_img_p],
+                        outputs=[segm_img_p, cond_img_p])
+
+    with gr.Tab("Text mode"):
+        # Images
+        with gr.Row(variant="panel"):
+            with gr.Column(scale=1):
+                cond_img_t.render()
+
+            with gr.Column(scale=1):
+                segm_img_t.render()
+
+        # Submit & Clear
+        with gr.Row():
+            with gr.Column():
+                input_size_slider_t = gr.components.Slider(minimum=512,
+                                                           maximum=1024,
+                                                           value=1024,
+                                                           step=64,
+                                                           label='Input_size',
+                                                           info='Our model was trained on a size of 1024')
+                with gr.Row():
+                    with gr.Column():
+                        contour_check = gr.Checkbox(value=True, label='withContours', info='draw the edges of the masks')
+                        text_box = gr.Textbox(label="text prompt", value="a black dog")
+
+                    with gr.Column():
+                        segment_btn_t = gr.Button("Segment with text", variant='primary')
+                        clear_btn_t = gr.Button("Clear", variant="secondary")
+
+                gr.Markdown("Try some of the examples below ⬇️")
+                gr.Examples(examples=[["examples/dogs.jpg"], ["examples/fruits.jpg"], ["examples/flowers.jpg"]],
+                            inputs=[cond_img_t],
+                            # outputs=segm_img_e,
+                            # fn=segment_everything,
+                            # cache_examples=True,
+                            examples_per_page=4)
+
+            with gr.Column():
+                with gr.Accordion("Advanced options", open=False):
+                    iou_threshold = gr.Slider(0.1, 0.9, 0.7, step=0.1, label='iou', info='iou threshold for filtering the annotations')
+                    conf_threshold = gr.Slider(0.1, 0.9, 0.25, step=0.05, label='conf', info='object confidence threshold')
+                    with gr.Row():
+                        mor_check = gr.Checkbox(value=False, label='better_visual_quality', info='better quality using morphologyEx')
+                        retina_check = gr.Checkbox(value=True, label='use_retina', info='draw high-resolution segmentation masks')
+                        wider_check = gr.Checkbox(value=False, label='wider', info='wider result')
+
+                # Description
+                gr.Markdown(description_e)
+    
+    segment_btn_t.click(segment_everything,
+                        inputs=[
+                            cond_img_t,
+                            input_size_slider_t,
+                            iou_threshold,
+                            conf_threshold,
+                            mor_check,
+                            contour_check,
+                            retina_check,
+                            text_box,
+                            wider_check,
+                        ],
+                        outputs=segm_img_t)
+
+    def clear():
+        return None, None
+    
+    def clear_text():
+        return None, None, None
+
+    clear_btn_e.click(clear, outputs=[cond_img_e, segm_img_e])
+    clear_btn_p.click(clear, outputs=[cond_img_p, segm_img_p])
+    clear_btn_t.click(clear_text, outputs=[cond_img_p, segm_img_p, text_box])
+
+demo.queue()
+demo.launch()

gitattributes.txt

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.jpg filter=lfs diff=lfs merge=lfs -text

gitignore.txt

@@ -0,0 +1,6 @@
+*.pyc
+*.pyo
+*.pyd
+.DS_Store
+.idea
+gradio_cached_examples

requirements.txt

@@ -0,0 +1,20 @@
+# Base-----------------------------------
+matplotlib==3.2.2
+numpy
+opencv-python
+
+git+https://github.com/openai/CLIP.git
+# Pillow>=7.1.2
+# PyYAML>=5.3.1
+# requests>=2.23.0
+# scipy>=1.4.1
+# torch
+# torchvision
+# tqdm>=4.64.0
+
+# pandas>=1.1.4
+# seaborn>=0.11.0
+
+# Ultralytics-----------------------------------
+ultralytics==8.0.121
+