AP-10K

Dataset Information
Modalities
Images
Introduced
2021
License
Unknown
Homepage
Official Website
Contents

Overview

AP-10K is the first large-scale benchmark for general animal pose estimation, to facilitate the research in animal pose estimation. AP-10K consists of 10,015 images collected and filtered from 23 animal families and 60 species following the taxonomic rank and high-quality keypoint annotations labeled and checked manually.

Source: AP-10K: A Benchmark for Animal Pose Estimation in the Wild

Image source: https://arxiv.org/pdf/2108.12617v1.pdf

Variants: AP-10K

Associated Benchmarks

This dataset is used in 2 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
Semantic correspondence DINOv2 Semantic Correspondence: Unified Benchmarking and … 2025-05-23
Animal Pose Estimation UniPose X-Pose: Detecting Any Keypoints 2023-10-12
Animal Pose Estimation ViTPose+-B ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation ViTPose+-H ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation HRNet-w32 ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation ViTPose+-S ViT-S ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation SimpleBaseline-ResNet50 ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation HRNet-w48 ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation ViTPose+-L ViTPose++: Vision Transformer for Generic … 2022-12-07
Animal Pose Estimation zero-shot SuperAnimal-HRNetw32 SuperAnimal pretrained pose estimation models … 2022-03-14
Animal Pose Estimation SuperAnimal-HRNetw32 SuperAnimal pretrained pose estimation models … 2022-03-14

Research Papers

Recent papers with results on this dataset:

External Links: