# LInKs "Lifting Independent Keypoints" -- Partial Pose Lifting for Occlusion Handling with Improved Accuracy in 2D-3D Human Pose Estimation

arXiv: https://arxiv.org/abs/2309.07243 · HTML: https://mlsanity.com/paper/2309.07243/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 33 references that are in the map.
- First posted on arXiv 13 Sep 2023.

## Builds on (most cited first, 21 of 21)

- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Density estimation using Real NVP](https://mlsanity.com/paper/1605.08803/) (2016)
- [OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields](https://mlsanity.com/paper/1812.08008/) (2018)
- [Stacked Hourglass Networks for Human Pose Estimation](https://mlsanity.com/paper/1603.06937/) (2016)
- [Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision](https://mlsanity.com/paper/1611.09813/) (2016)
- [3D human pose estimation in video with temporal convolutions and semi-supervised training](https://mlsanity.com/paper/1811.11742/) (2018)
- [A simple yet effective baseline for 3d human pose estimation](https://mlsanity.com/paper/1705.03098/) (2017)
- [Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose](https://mlsanity.com/paper/1611.07828/) (2016)
- [Unsupervised 3D Pose Estimation with Geometric Self-Supervision](https://mlsanity.com/paper/1904.04812/) (2019)
- [3D Human Pose Estimation in the Wild by Adversarial Learning](https://mlsanity.com/paper/1803.09722/) (2018)
- [RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation](https://mlsanity.com/paper/1902.09868/) (2019)
- [Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image](https://mlsanity.com/paper/1701.00295/) (2017)
- [Can 3D Pose be Learned from 2D Projections Alone?](https://mlsanity.com/paper/1808.07182/) (2018)
- [Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis](https://mlsanity.com/paper/2004.04400/) (2020)
- [Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimation](https://mlsanity.com/paper/1508.06708/) (2015)
- [ElePose: Unsupervised 3D Human Pose Estimation by Predicting Camera Elevation and Learning Normalizing Flows on 2D Poses](https://mlsanity.com/paper/2112.07088/) (2021)
- [Adversarial Inverse Graphics Networks: Learning 2D-to-3D Lifting and Image-to-Image Translation from Unpaired Supervision](https://mlsanity.com/paper/1705.11166/) (2017)
- [Unsupervised Adversarial Learning of 3D Human Pose from 2D Joint Locations](https://mlsanity.com/paper/1803.08244/) (2018)
- [3D Human Pose Estimation on a Configurable Bed from a Pressure Image](https://mlsanity.com/paper/1804.07873/) (2018)
- [AnimePose: Multi-person 3D pose estimation and animation](https://mlsanity.com/paper/2002.02792/) (2020)
- [A Kinematic Chain Space for Monocular Motion Capture](https://mlsanity.com/paper/1702.00186/) (2017)

## Built on by (most cited first, 6 of 6)

- [Benchmarking 3D Human Pose Estimation Models under Occlusions](https://mlsanity.com/paper/2504.10350/) (2025)
- [Improving Real-Time Omnidirectional 3D Multi-Person Human Pose Estimation with People Matching and Unsupervised 2D-3D Lifting](https://mlsanity.com/paper/2403.09437/) (2024)
- [Unsupervised Multi-Person 3D Human Pose Estimation From 2D Poses Alone](https://mlsanity.com/paper/2309.14865/) (2023)
- [L3D-Pose: Lifting Pose for 3D Avatars from a Single Camera in the Wild](https://mlsanity.com/paper/2501.01174/) (2025)
- [Two Views Are Better than One: Monocular 3D Pose Estimation with Multiview Consistency](https://mlsanity.com/paper/2311.12421/) (2023)
- [LaxMotion: Rethinking Supervision Granularity for 3D Human Motion Generation](https://mlsanity.com/paper/2511.11368/) (2025)

## Abstract (arXiv)

We present LInKs, a novel unsupervised learning method to recover 3D human poses from 2D kinematic skeletons obtained from a single image, even when occlusions are present. Our approach follows a unique two-step process, which involves first lifting the occluded 2D pose to the 3D domain, followed by filling in the occluded parts using the partially reconstructed 3D coordinates. This lift-then-fill approach leads to significantly more accurate results compared to models that complete the pose in 2D space alone. Additionally, we improve the stability and likelihood estimation of normalising flows through a custom sampling function replacing PCA dimensionality reduction previously used in prior work. Furthermore, we are the first to investigate if different parts of the 2D kinematic skeleton can be lifted independently which we find by itself reduces the error of current lifting approaches. We attribute this to the reduction of long-range keypoint correlations. In our detailed evaluation, we quantify the error under various realistic occlusion scenarios, showcasing the versatility and applicability of our model. Our results consistently demonstrate the superiority of handling all types of occlusions in 3D space when compared to others that complete the pose in 2D space. Our approach also exhibits consistent accuracy in scenarios without occlusion, as evidenced by a 7.9% reduction in reconstruction error compared to prior works on the Human3.6M dataset. Furthermore, our method excels in accurately retrieving complete 3D poses even in the presence of occlusions, making it highly applicable in situations where complete 2D pose information is unavailable.