Implementation:Open compass VLMEvalKit MMGenBench

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, Image Generation Prompt

Overview

Benchmark dataset implementation for MMGenBench evaluation in VLMEvalKit.

Description

MMGenBench inherits from ImageBaseDataset and implements the MMGenBench benchmark for evaluating image-to-caption-prompt generation. The TYPE field is set to 'GenerateImgPrompt'. It generates image caption prompts suitable for text-to-image models like Stable Diffusion.

Usage

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

Code Reference

Source: vlmeval/dataset/mmgenbench.py, Lines: L1-69
Import: from vlmeval.dataset.mmgenbench import MMGenBench

Signature:

class MMGenBench(ImageBaseDataset):
    TYPE = 'GenerateImgPrompt'
    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('MMGenBench-Test')

Related Pages

Page Connections

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

Principle

Implementation

Heuristic

Environment