Llava Video 7B Qwen2

🚀 LLaVA-Video-7B-Qwen2

LLaVA-Video-7B-Qwen2 是基於 Qwen2 語言模型的多模態模型，在圖像、多圖像和視頻交互方面表現出色，尤其專注於視頻處理。它在多個多模態數據集上進行了訓練和測試，具有較高的準確性。

🚀 快速開始

模型概述

LLaVA-Video 系列模型是參數為 7/72B 的模型，在 LLaVA-Video-178K 和 LLaVA-OneVision 數據集 上進行訓練，基於 Qwen2 語言模型，上下文窗口為 32K 個標記。

此模型最多支持 64 幀。

• 項目頁面：項目頁面
• 論文：更多詳情請查看我們的 論文
• 代碼倉庫：LLaVA-VL/LLaVA-NeXT
• 聯繫人：張元瀚
• 支持語言：英語、中文

模型使用

預期用途

該模型在 LLaVA-Video-178K 和 LLaVA-OneVision 數據集 上進行訓練，具備與圖像、多圖像和視頻進行交互的能力，尤其專注於視頻處理。

歡迎在社區板塊分享你的生成結果！

生成示例

我們提供了使用該模型的簡單生成流程。更多詳情可參考 Github。

# 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(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 = "lmms-lab/LLaVA-Video-7B-Qwen2"
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 = "XXXX"
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().half()
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"\n{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)

🔧 技術細節

訓練信息

模型

• 架構：SO400M + Qwen2
• 初始化模型：lmms-lab/llava-onevision-qwen2-7b-si
• 訓練數據：160 萬個單圖像/多圖像/視頻數據的混合，1 個訓練週期，全量模型訓練
• 精度：bfloat16

硬件與軟件

• GPU：256 塊英偉達 Tesla A100（用於整個模型系列的訓練）
• 編排工具：Huggingface Trainer
• 神經網絡框架：PyTorch

評估指標

📄 許可證

本項目採用 Apache-2.0 許可證。

📖 引用

@misc{zhang2024videoinstructiontuningsynthetic,
    title={Video Instruction Tuning With Synthetic Data}, 
    author={Yuanhan Zhang and Jinming Wu and Wei Li and Bo Li and Zejun Ma and Ziwei Liu and Chunyuan Li},
    year={2024},
    eprint={2410.02713},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2410.02713}, 
}