Implementation:Open compass VLMEvalKit VisFactor

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, Visual Factor Analysis

Overview

Benchmark dataset implementation for VisFactor evaluation in VLMEvalKit.

Description

VisFactor inherits from ImageBaseDataset and implements the VisFactor benchmark for visual factor analysis evaluation. The TYPE field is set to 'VQA'. It supports multiple variants including standard, CoT, and specialized GE/GN/GH splits.

Usage

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

Code Reference

Source: vlmeval/dataset/visfactor.py, Lines: L1-152
Import: from vlmeval.dataset.visfactor import VisFactor

Signature:

class VisFactor(ImageBaseDataset):
    TYPE = '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('VisFactor')

Related Pages

Page Connections

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

Principle

Implementation

Heuristic

Environment