# View-Invariant, Occlusion-Robust Probabilistic Embedding for Human Pose

arXiv: https://arxiv.org/abs/2010.13321 · HTML: https://mlsanity.com/paper/2010.13321/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Oct 2020.

## Built on by (most cited first, 9 of 9)

- [P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation](https://mlsanity.com/paper/2203.07628/) (2022)
- [SIMONe: View-Invariant, Temporally-Abstracted Object Representations via Unsupervised Video Decomposition](https://mlsanity.com/paper/2106.03849/) (2021)
- [Learning from Semantic Alignment between Unpaired Multiviews for Egocentric Video Recognition](https://mlsanity.com/paper/2308.11489/) (2023)
- [CHROME: Clothed Human Reconstruction with Occlusion-Resilience and Multiview-Consistency from a Single Image](https://mlsanity.com/paper/2503.15671/) (2025)
- [LDTrack: Dynamic People Tracking by Service Robots using Diffusion Models](https://mlsanity.com/paper/2402.08774/) (2024)
- [A Light-Weight Contrastive Approach for Aligning Human Pose Sequences](https://mlsanity.com/paper/2303.04244/) (2023)
- [SHARE: Single-view Human Adversarial REconstruction](https://mlsanity.com/paper/2401.00343/) (2023)
- [VIFSS: View-Invariant and Figure Skating-Specific Pose Representation Learning for Temporal Action Segmentation](https://mlsanity.com/paper/2508.10281/) (2025)
- [A Study on Real-time Object Detection using Deep Learning](https://mlsanity.com/paper/2602.15926/) (2026)

## Abstract (arXiv)

Recognition of human poses and actions is crucial for autonomous systems to interact smoothly with people. However, cameras generally capture human poses in 2D as images and videos, which can have significant appearance variations across viewpoints that make the recognition tasks challenging. To address this, we explore recognizing similarity in 3D human body poses from 2D information, which has not been well-studied in existing works. Here, we propose an approach to learning a compact view-invariant embedding space from 2D body joint keypoints, without explicitly predicting 3D poses. Input ambiguities of 2D poses from projection and occlusion are difficult to represent through a deterministic mapping, and therefore we adopt a probabilistic formulation for our embedding space. Experimental results show that our embedding model achieves higher accuracy when retrieving similar poses across different camera views, in comparison with 3D pose estimation models. We also show that by training a simple temporal embedding model, we achieve superior performance on pose sequence retrieval and largely reduce the embedding dimension from stacking frame-based embeddings for efficient large-scale retrieval. Furthermore, in order to enable our embeddings to work with partially visible input, we further investigate different keypoint occlusion augmentation strategies during training. We demonstrate that these occlusion augmentations significantly improve retrieval performance on partial 2D input poses. Results on action recognition and video alignment demonstrate that using our embeddings without any additional training achieves competitive performance relative to other models specifically trained for each task.