Implementation:Open compass VLMEvalKit MMBenchVideo

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, Video VQA

Overview

Benchmark dataset implementation for MMBench-Video evaluation in VLMEvalKit.

Description

MMBenchVideo inherits from VideoBaseDataset and implements the MMBench-Video benchmark for video visual question answering. The TYPE field is set to 'Video-VQA'. It supports pack and no-pack modes with different prompt templates for single and multiple questions.

Usage

Registered in vlmeval/dataset/__init__.py and invoked through build_dataset() by benchmark name.

Code Reference

Source: vlmeval/dataset/mmbench_video.py, Lines: L1-257
Import: from vlmeval.dataset.mmbench_video import MMBenchVideo

Signature:

class MMBenchVideo(VideoBaseDataset):
    TYPE = 'Video-VQA'
    DATASET_URL = {...}
    DATASET_MD5 = {...}
    ...

I/O Contract

Direction	Description
Inputs	TSV dataset file with image/video paths and questions
Outputs	Evaluation results DataFrame with scores per category

Usage Examples

from vlmeval.dataset import build_dataset
dataset = build_dataset('MMBench-Video')

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment