Hugging Face's logo

Vision-CAIR
/
LongVU_Qwen2_7B

Video-Text-to-Text
Safetensors
cambrian_qwen
Eval Results
Model card Files Community
8
LongVU_Qwen2_7B
File size: 3,264 Bytes 
5b15eaf
 
 
 
 
d69cfa6
5b15eaf
d69cfa6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5b15eaf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
 
---
datasets:
- shenxq/OneVision
- shenxq/VideoChat2
base_model:
- Vision-CAIR/LongVU_Qwen2_7B_img
model-index:
- name: llava-onevision-qwen-7b-ov
  results:
  - task:
      type: multimodal
    dataset:
      name: EgoSchema
      type: egoschema
    metrics:
    - type: accuracy
      value: 67.6
      name: accuracy
      verified: true
  - task:
      type: multimodal
    dataset:
      name: MLVU
      type: mlvu
    metrics:
    - type: accuracy
      value: 65.4
      name: accuracy
      verified: true
  - task:
      type: multimodal
    dataset:
      name: MVBench
      type: mvbench
    metrics:
    - type: accuracy
      value: 66.9
      name: accuracy
      verified: true
  - task:
      type: multimodal
    dataset:
      name: VideoMME
      type: videomme
    metrics:
    - type: accuracy
      value: 60.6
      name: accuracy
      verified: true
---
# LongVU

Play with the model on the [HF demo](https://huggingface.co/spaces/Vision-CAIR/LongVU).

<div align="left">
    <a href='https://vision-cair.github.io/LongVU'><img src="https://longvu.s3.amazonaws.com/assets/demo.gif" alt="Demo GIF" style="width: 100%; max-width: 650px;"></a>
</div>

# Use

We provide the simple generation process for using our model. For more details, you could refer to [Github](https://github.com/Vision-CAIR/LongVU)

```python
# git clone https://github.com/Vision-CAIR/LongVU
import numpy as np
import torch
from longvu.builder import load_pretrained_model
from longvu.constants import (
    DEFAULT_IMAGE_TOKEN,
    IMAGE_TOKEN_INDEX,
)
from longvu.conversation import conv_templates, SeparatorStyle
from longvu.mm_datautils import (
    KeywordsStoppingCriteria,
    process_images,
    tokenizer_image_token,
)
from decord import cpu, VideoReader

tokenizer, model, image_processor, context_len = load_pretrained_model(
    "./checkpoints/longvu_qwen", None, "cambrian_qwen",
)

model.eval()
video_path = "./examples/video1.mp4"
qs = "Describe this video in detail"

vr = VideoReader(video_path, ctx=cpu(0), num_threads=1)
fps = float(vr.get_avg_fps())
frame_indices = np.array([i for i in range(0, len(vr), round(fps),)])
video = []
for frame_index in frame_indices:
    img = vr[frame_index].asnumpy()
    video.append(img)
video = np.stack(video)
image_sizes = [video[0].shape[:2]]
video = process_images(video, image_processor, model.config)
video = [item.unsqueeze(0) for item in video]

qs = DEFAULT_IMAGE_TOKEN + "\n" + qs
conv = conv_templates["qwen"].copy()
conv.append_message(conv.roles[0], qs)
conv.append_message(conv.roles[1], None)
prompt = conv.get_prompt()

input_ids = tokenizer_image_token(prompt, tokenizer, IMAGE_TOKEN_INDEX, return_tensors="pt").unsqueeze(0).to(model.device)
stop_str = conv.sep if conv.sep_style != SeparatorStyle.TWO else conv.sep2
keywords = [stop_str]
stopping_criteria = KeywordsStoppingCriteria(keywords, tokenizer, input_ids)
with torch.inference_mode():
    output_ids = model.generate(
        input_ids,
        images=video,
        image_sizes=image_sizes,
        do_sample=False,
        temperature=0.2,
        max_new_tokens=128,
        use_cache=True,
        stopping_criteria=[stopping_criteria],
    )
pred = tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0].strip()
```