LSP (Leeds Sports Pose)

Introduced in Clustered Pose and Nonlinear Appearance Models for Human Pose Estimation1 Jan 2010 archive 2025-07-28

The Leeds Sports Pose (LSP) dataset is widely used as the benchmark for human pose estimation. The original LSP dataset contains 2,000 images of sportspersons gathered from Flickr, 1000 for training and 1000 for testing. Each image is annotated with 14 joint locations, where left and right joints are consistently labelled from a person-centric viewpoint. The extended LSP dataset contains additional 10,000 images labeled for training.

Source: Deep Deformation Network for Object Landmark Localization

Image: Sumer et al

Homepage Benchmarks Papers JSON

Benchmarks archive 2025-07-28

All 1 leaderboard whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

		First row (archive order)	Paper	Code
Pose Estimation	Leeds Sports Poses	OmniPose PCK 99.5%	OmniPose: A Multi-Scale Framework for Multi-Person Pose...	bmartacho/OmniPose	18	Compare

Papers archive 2025-07-28

18 shown of 18 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 202. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

			Date	Samples run Syntology
Learning Dynamical Human-Joint Affinity for 3D Pose Estimation in Videos	0	1	15 Sep 2021	not harvested
OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation	1	1	18 Mar 2021	not harvested
Toward fast and accurate human pose estimation via soft-gated skip connections	3	1	25 Feb 2020	not harvested
UniPose: Unified Human Pose Estimation in Single Images and Videos	2	1	22 Jan 2020	not harvested
Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation	1	1	22 Aug 2019	not harvested
Fast Human Pose Estimation	1	1	13 Nov 2018	not harvested
Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation	0	1	24 May 2018	not harvested
Human Pose Regression by Combining Indirect Part Detection and Contextual Information	1	1	6 Oct 2017	ran 0 of 1 samples (1 unverified)
Learning Feature Pyramids for Human Pose Estimation	3	1	3 Aug 2017	not harvested
Self Adversarial Training for Human Pose Estimation	1	1	8 Jul 2017	not harvested
Knowledge-Guided Deep Fractal Neural Networks for Human Pose Estimation	1	1	5 May 2017	not harvested
VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera	1	1	3 May 2017	not harvested
Multi-Context Attention for Human Pose Estimation	2	1	24 Feb 2017	not harvested
Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision	0	1	29 Nov 2016	not harvested
Human pose estimation via Convolutional Part Heatmap Regression	1	1	6 Sep 2016	not harvested
DeeperCut: A Deeper, Stronger, and Faster Multi-Person Pose Estimation Model	16	1	10 May 2016	not harvested
Convolutional Pose Machines	50	1	30 Jan 2016	ran 1 of 4 samples (3 unverified; 3 pointer-only for licence)
Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations	0	1	12 Jul 2014	not harvested

Dataset loaders archive 2025-07-28

docs.activeloop.ai/datasets/lsp-dataset activeloopai/Hubtfpytorch

open-mmlab/mmpose open-mmlab/mmposepytorch

gas.graviti.com/dataset/graviti/LeedsSportsPose Graviti-AI/datasetstfpytorch

3 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Pose Estimation 3D Human Pose Estimation 3D Pose Estimation

License archive 2025-07-28

Custom

Modalities archive 2025-07-28

Images

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

Leeds Sports Poses
Leeds Sports Pose
LSP

3 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections