Implementation:Open compass VLMEvalKit UniSVG

Field	Value
source	VLMEvalKit
domain	Vision, Benchmarking, SVG Understanding

Overview

Benchmark dataset implementation for UniSVG evaluation in VLMEvalKit.

Description

UniSVG inherits from ImageBaseDataset and implements the UniSVG benchmark for unified SVG understanding evaluation. The TYPE field is set to 'VQA'. It supports multiple task types including ISVGEN, ISVGUN (usage, category, description, color) for image-based tasks.

Usage

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

Code Reference

Source: vlmeval/dataset/uni_svg.py, Lines: L1-51
Import: from vlmeval.dataset.uni_svg import UniSVG

Signature:

class UniSVG(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('UniSVG')

Related Pages

Page Connections

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

Principle

Implementation

Heuristic

Environment