Visual Evaluation with Foundation Models

Visual Evaluation with Foundation Models's repositories

Q-Align

③[ICML2024] [IQA, IAA, VQA] All-in-one Foundation Model for visual scoring. Can efficiently fine-tune to downstream datasets.

Language:PythonNOASSERTION517 3 45

Q-Bench

①[ICLR2024 Spotlight] (GPT-4V/Gemini-Pro/Qwen-VL-Plus+16 OS MLLMs) A benchmark for multi-modality LLMs (MLLMs) on low-level vision and visual quality assessment.

Language:Jupyter NotebookNOASSERTION279 1 12

Q-Instruct

②[CVPR 2024] Low-level visual instruction tuning, with a 200K dataset and a model zoo for fine-tuned checkpoints.

Language:PythonNOASSERTION230 2 28

A-Bench

[ICLR 2025] What do we expect from LMMs as AIGI evaluators and how do they perform?

139 50

Q-Bench-Video

[CVPR 2025] A benchmark for video quality understanding of LMMs

Language:Python109 1 1

Co-Instruct

④[ECCV 2024 Oral, Comparison among Multiple Images!] A study on open-ended multi-image quality comparison: a dataset, a model and a benchmark.

NOASSERTION86 1 7

Visual-Question-Answering-for-Video-Quality-Assessment

[ACMMM2025] Official released code for VQA² series models

Language:PythonApache-2.05000

Compare2Score

[Neurips 24 Spotlight] Training in Pairs + Inference on Single Image with Anchors

Language:PythonMIT41 2 6

Q-Ground

Official codes for "Q-Ground: Image Quality Grounding with Large Multi-modality Models", ACM MM2024 (Oral)

NOASSERTION38 3 3

CMC-Bench

[ACMMM 2025] Benchmarking MLLM Codec Ability

Language:PythonMIT32 10

Q-Refine

[MM 2024 Oral] Refiner for AIGC

Language:Jupyter NotebookApache-2.027 2 2

Chinese-Q-Bench

[WIP@Oct 13] 质衡-基准测试 (Q-Bench in Chinese)，包含中文版【底层视觉问答】和【底层视觉描述】数据集，以及中文提示下的图片质量评价。 We will release Q-Bench in more languages in the future.

2202

LMM-PCQA

Official repo for `LMM-PCQA: Assisting Point Cloud Quality Assessment with LMM', ACM MM2024 Oral

Language:Python16 1 1

R-Bench

[IEEE JSTSP 2025] Using LMM in Real-world Distortion

Language:Jupyter Notebook9 1 1

.github

We are an open-source collaborative project to bring new possibilities to IQA!

200

Q-Align-Aria

Visual Scorer with Aria MoE model

Language:Jupyter NotebookApache-2.0200