Hugging Face's logo

Datasets:
rxc205
/
EPIC-Bench

Tasks:
Visual Question Answering
Object Detection
Modalities:
Image
Text
Formats:
parquet
Languages:
English
Size:
1K - 10K
ArXiv:
Tags:
embodied-perception
mask-grounding
vision-language-models
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
rxc205 commited on
Commit
d9321bc
·
verified ·
1 Parent(s): 7450c2d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -5
README.md CHANGED
@@ -4,7 +4,6 @@ pretty_name: EPIC-Bench
4
  task_categories:
5
  - visual-question-answering
6
  - visual-grounding
7
- - image-segmentation
8
  language:
9
  - en
10
  Modalities:
@@ -30,23 +29,23 @@ configs:
30
 
31
  [![arXiv](https://img.shields.io/badge/arXiv-coming_soon-b31b1b.svg)](https://epic-bench.github.io/EPIC-Bench/)
32
  [![Project Page](https://img.shields.io/badge/Project-Page-blue)](https://epic-bench.github.io/EPIC-Bench/)
33
- [![Dataset](https://img.shields.io/badge/🤗-Dataset-yellow)](https://epic-bench.github.io/EPIC-Bench/)
34
  [![Evaluation Toolkit](https://img.shields.io/badge/⚙️-Evaluation_Toolkit-6366f1.svg)](#-epic-bench-evaluation-toolkit)
35
  [![License](https://img.shields.io/badge/License-TBD-lightgrey.svg)](#-license)
36
 
37
- [**Homepage**](https://epic-bench.github.io/EPIC-Bench/) | [**Paper**](https://epic-bench.github.io/EPIC-Bench/) | [**Dataset**](https://epic-bench.github.io/EPIC-Bench/) | [**Leaderboard**](https://epic-bench.github.io/EPIC-Bench/)
38
 
39
  </div>
40
 
41
  ## 📃 Overview
42
 
43
  > This repo contains the official evaluation code and dataset for the paper
44
- > **"EPIC-Bench: Can VLMs Perceive the Embodied Real-World?"**
45
 
46
  **EPIC-Bench** is a **Mask-Grounding-based** benchmark designed to evaluate a VLM’s **Visual Perception** capability in **Embodied Scenarios**.
47
 
48
  <p align="center">
49
- <img src="https://github.com/wei0623kb/EPIC-Bench-Eval/blob/main/images/teaser.png?raw=true" alt="EPIC-Bench teaser" width="100%"/>
50
  </p>
51
 
52
  📚 EPIC-Bench covers **3 High-Level Categories** and **23 Task Types**, following the realistic **Embodied Workflow**:
 
4
  task_categories:
5
  - visual-question-answering
6
  - visual-grounding
 
7
  language:
8
  - en
9
  Modalities:
 
29
 
30
  [![arXiv](https://img.shields.io/badge/arXiv-coming_soon-b31b1b.svg)](https://epic-bench.github.io/EPIC-Bench/)
31
  [![Project Page](https://img.shields.io/badge/Project-Page-blue)](https://epic-bench.github.io/EPIC-Bench/)
32
+ [![Dataset](https://img.shields.io/badge/🤗-Dataset-yellow)](https://huggingface.co/datasets/rxc205/EPIC-Bench)
33
  [![Evaluation Toolkit](https://img.shields.io/badge/⚙️-Evaluation_Toolkit-6366f1.svg)](#-epic-bench-evaluation-toolkit)
34
  [![License](https://img.shields.io/badge/License-TBD-lightgrey.svg)](#-license)
35
 
36
+ [**Homepage**](https://epic-bench.github.io/EPIC-Bench/) | [**Paper**](https://epic-bench.github.io/EPIC-Bench/) | [**Dataset**](https://huggingface.co/datasets/rxc205/EPIC-Bench) | [**Leaderboard**](https://epic-bench.github.io/EPIC-Bench/#leaderboard-section)
37
 
38
  </div>
39
 
40
  ## 📃 Overview
41
 
42
  > This repo contains the official evaluation code and dataset for the paper
43
+ > **"EPIC-Bench: A Perception-Centric Benchmark for Fine-Grained Embodied Visual Grounding in Vision-Language Models"**
44
 
45
  **EPIC-Bench** is a **Mask-Grounding-based** benchmark designed to evaluate a VLM’s **Visual Perception** capability in **Embodied Scenarios**.
46
 
47
  <p align="center">
48
+ <img src="https://raw.githubusercontent.com/rxc205/EPIC-Bench-Eval/refs/heads/main/images/teaser.png" alt="EPIC-Bench teaser" width="100%"/>
49
  </p>
50
 
51
  📚 EPIC-Bench covers **3 High-Level Categories** and **23 Task Types**, following the realistic **Embodied Workflow**: