# SCJD: Sparse Correlation and Joint Distillation for Efficient 3D Human Pose Estimation

arXiv: https://arxiv.org/abs/2503.14097 · HTML: https://mlsanity.com/paper/2503.14097/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 44 references that are in the map.
- First posted on arXiv 18 Mar 2025.

## Builds on (most cited first, 28 of 28)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Deep High-Resolution Representation Learning for Human Pose Estimation](https://mlsanity.com/paper/1902.09212/) (2019)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [Similarity-Preserving Knowledge Distillation](https://mlsanity.com/paper/1907.09682/) (2019)
- [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)
- [Decoupled Knowledge Distillation](https://mlsanity.com/paper/2203.08679/) (2022)
- [Cascaded Pyramid Network for Multi-Person Pose Estimation](https://mlsanity.com/paper/1711.07319/) (2017)
- [3D Human Pose Estimation with Spatial and Temporal Transformers](https://mlsanity.com/paper/2103.10455/) (2021)
- [Learning Deep Representations with Probabilistic Knowledge Transfer](https://mlsanity.com/paper/1803.10837/) (2018)
- [MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video](https://mlsanity.com/paper/2203.00859/) (2022)
- [MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2111.12707/) (2021)
- [Knowledge Distillation with the Reused Teacher Classifier](https://mlsanity.com/paper/2203.14001/) (2022)
- [Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2103.14304/) (2021)
- [PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation](https://mlsanity.com/paper/2303.17472/) (2023)
- [Self-supervised Knowledge Distillation Using Singular Value Decomposition](https://mlsanity.com/paper/1807.06819/) (2018)
- [KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2404.00658/) (2024)
- [Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformers](https://mlsanity.com/paper/2210.06110/) (2022)
- [On the Benefits of 3D Pose and Tracking for Human Action Recognition](https://mlsanity.com/paper/2304.01199/) (2023)
- [Beat-It: Beat-Synchronized Multi-Condition 3D Dance Generation](https://mlsanity.com/paper/2407.07554/) (2024)
- [HDFormer: High-order Directed Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2302.01825/) (2023)
- [G2Face: High-Fidelity Reversible Face Anonymization via Generative and Geometric Priors](https://mlsanity.com/paper/2408.09458/) (2024)
- [CrossFormer: Cross Spatio-Temporal Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2203.13387/) (2022)
- [Lightweight 3D Human Pose Estimation Network Training Using Teacher-Student Learning](https://mlsanity.com/paper/2001.05097/) (2020)
- [Rotation-Adaptive Point Cloud Domain Generalization via Intricate Orientation Learning](https://mlsanity.com/paper/2502.02247/) (2025)

## Built on by (most cited first, 2 of 2)

- [ViewSRD: 3D Visual Grounding via Structured Multi-View Decomposition](https://mlsanity.com/paper/2507.11261/) (2025)
- [StarPose: 3D Human Pose Estimation via Spatial-Temporal Autoregressive Diffusion](https://mlsanity.com/paper/2508.02056/) (2025)

## Abstract (arXiv)

Existing 3D Human Pose Estimation (HPE) methods achieve high accuracy but suffer from computational overhead and slow inference, while knowledge distillation methods fail to address spatial relationships between joints and temporal correlations in multi-frame inputs. In this paper, we propose Sparse Correlation and Joint Distillation (SCJD), a novel framework that balances efficiency and accuracy for 3D HPE. SCJD introduces Sparse Correlation Input Sequence Downsampling to reduce redundancy in student network inputs while preserving inter-frame correlations. For effective knowledge transfer, we propose Dynamic Joint Spatial Attention Distillation, which includes Dynamic Joint Embedding Distillation to enhance the student's feature representation using the teacher's multi-frame context feature, and Adjacent Joint Attention Distillation to improve the student network's focus on adjacent joint relationships for better spatial understanding. Additionally, Temporal Consistency Distillation aligns the temporal correlations between teacher and student networks through upsampling and global supervision. Extensive experiments demonstrate that SCJD achieves state-of-the-art performance. Code is available at https://github.com/wileychan/SCJD.