Llava Video 7B Qwen2 TPO

🚀 LLaVA-Video-7B-Qwen2-TPO

LLaVA-Video-7B-Qwen2-TPO是基於論文Temporal Preference Optimization for Long-form Video Understanding提出的模型，它在LLaVA-Video-7B-Qwen2的基礎上進行了時間偏好優化。該模型在一系列基準測試中取得了領先的性能，與LLaVA-Video-7B相比，平均性能提升了1.5%。值得注意的是，它在Video-MME基準測試中成為了領先的70億參數模型。

項目頁面：https://ruili33.github.io/tpo_website/ 代碼：https://github.com/ruili33/TPO

🚀 快速開始

使用以下代碼開始使用該模型。更多信息，請參考我們的GitHub倉庫。

基礎用法

# pip install git+https://github.com/LLaVA-VL/LLaVA-NeXT.git
from llava.model.builder import load_pretrained_model
from llava.mm_utils import get_model_name_from_path, process_images, tokenizer_image_token
from llava.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN, DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN, IGNORE_INDEX
from llava.conversation import conv_templates, SeparatorStyle
from PIL import Image
import requests
import copy
import torch
import sys
import warnings
from decord import VideoReader, cpu
import numpy as np
warnings.filterwarnings("ignore")
def load_video(self, video_path, max_frames_num,fps=1,force_sample=False):
    if max_frames_num == 0:
        return np.zeros((1, 336, 336, 3))
    vr = VideoReader(video_path, ctx=cpu(0),num_threads=1)
    total_frame_num = len(vr)
    video_time = total_frame_num / vr.get_avg_fps()
    fps = round(vr.get_avg_fps()/fps)
    frame_idx = [i for i in range(0, len(vr), fps)]
    frame_time = [i/fps for i in frame_idx]
    if len(frame_idx) > max_frames_num or force_sample:
        sample_fps = max_frames_num
        uniform_sampled_frames = np.linspace(0, total_frame_num - 1, sample_fps, dtype=int)
        frame_idx = uniform_sampled_frames.tolist()
        frame_time = [i/vr.get_avg_fps() for i in frame_idx]
    frame_time = ",".join([f"{i:.2f}s" for i in frame_time])
    spare_frames = vr.get_batch(frame_idx).asnumpy()
    # import pdb;pdb.set_trace()
    return spare_frames,frame_time,video_time
pretrained = "ruili0/LLaVA-Video-7B-TPO"
model_name = "llava_qwen"
device = "cuda"
device_map = "auto"
tokenizer, model, image_processor, max_length = load_pretrained_model(pretrained, None, model_name, torch_dtype="bfloat16", device_map=device_map)  # Add any other thing you want to pass in llava_model_args
model.eval()
video_path = "local_demo/assets/dc_demo.mp4"
max_frames_num = "64"
video,frame_time,video_time = load_video(video_path, max_frames_num, 1, force_sample=True)
video = image_processor.preprocess(video, return_tensors="pt")["pixel_values"].cuda().bfloat16()
video = [video]
conv_template = "qwen_1_5"  # Make sure you use correct chat template for different models
time_instruciton = f"The video lasts for {video_time:.2f} seconds, and {len(video[0])} frames are uniformly sampled from it. These frames are located at {frame_time}.Please answer the following questions related to this video."
question = DEFAULT_IMAGE_TOKEN + f"{time_instruciton}\nPlease describe this video in detail."
conv = copy.deepcopy(conv_templates[conv_template])
conv.append_message(conv.roles[0], question)
conv.append_message(conv.roles[1], None)
prompt_question = conv.get_prompt()
input_ids = tokenizer_image_token(prompt_question, tokenizer, IMAGE_TOKEN_INDEX, return_tensors="pt").unsqueeze(0).to(device)
cont = model.generate(
    input_ids,
    images=video,
    modalities= ["video"],
    do_sample=False,
    temperature=0,
    max_new_tokens=4096,
)
text_outputs = tokenizer.batch_decode(cont, skip_special_tokens=True)[0].strip()
print(text_outputs)

✨ 主要特性

• LLaVA-Video-7B-Qwen2-TPO在LLaVA-Video-7B-Qwen2的基礎上，通過時間偏好進行優化。
• 該模型在一系列基準測試中取得了領先的性能，與LLaVA-Video-7B相比，平均性能提升了1.5%。
• 在Video-MME基準測試中，它成為了領先的70億參數模型。

📚 詳細文檔

評估結果

📄 許可證

本項目使用了某些數據集和檢查點，這些數據集和檢查點受其各自的原始許可證約束。用戶必須遵守這些原始許可證的所有條款和條件，包括但不限於數據集的OpenAI使用條款和基礎語言模型（Qwen2許可證）的特定許可證。本項目不會在原始許可證規定的範圍之外施加任何額外的限制。此外，提醒用戶確保其對數據集和檢查點的使用符合所有適用的法律法規。

🔖 引用

BibTeX引用

@article{li2025temporal,
      title={Temporal Preference Optimization for Long-Form Video Understanding},
      author={Li, Rui and Wang, Xiaohan and Zhang, Yuhui and Wang, Zeyu and Yeung-Levy, Serena},
      journal={arXiv preprint arXiv:2501.13919},
      year={2025}
    }

參考文獻

[1]. Liu, Z., Zhu, L., Shi, B., Zhang, Z., Lou, Y., Yang, S., ... & Lu, Y. (2024). NVILA: Efficient Frontier Visual Language Models. arXiv preprint arXiv:2412.04468.

[2]. Zhang, Y., Wu, J., Li, W., Li, B., Ma, Z., Liu, Z., & Li, C. (2024). Video instruction tuning with synthetic data. arXiv preprint arXiv:2410.02713.