# CrossFormer: Cross Spatio-Temporal Transformer for 3D Human Pose Estimation

arXiv: https://arxiv.org/abs/2203.13387 · HTML: https://mlsanity.com/paper/2203.13387/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 49 references that are in the map.
- First posted on arXiv 24 Mar 2022.

## Builds on (most cited first, 30 of 31)

- [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)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision](https://mlsanity.com/paper/1611.09813/) (2016)
- [End-to-End Human Pose and Mesh Reconstruction with Transformers](https://mlsanity.com/paper/2012.09760/) (2020)
- [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)
- [Cascaded Pyramid Network for Multi-Person Pose Estimation](https://mlsanity.com/paper/1711.07319/) (2017)
- [I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image](https://mlsanity.com/paper/2008.03713/) (2020)
- [3D Human Pose Estimation with Spatial and Temporal Transformers](https://mlsanity.com/paper/2103.10455/) (2021)
- [LocalViT: Analyzing Locality in Vision Transformers](https://mlsanity.com/paper/2104.05707/) (2021)
- [Semantic Graph Convolutional Networks for 3D Human Pose Regression](https://mlsanity.com/paper/1904.03345/) (2019)
- [VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera](https://mlsanity.com/paper/1705.01583/) (2017)
- [RegionViT: Regional-to-Local Attention for Vision Transformers](https://mlsanity.com/paper/2106.02689/) (2021)
- [Ordinal Depth Supervision for 3D Human Pose Estimation](https://mlsanity.com/paper/1805.04095/) (2018)
- [Graph Stacked Hourglass Networks for 3D Human Pose Estimation](https://mlsanity.com/paper/2103.16385/) (2021)
- [SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach](https://mlsanity.com/paper/2007.09389/) (2020)
- [Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2103.14304/) (2021)
- [Exploiting temporal information for 3D pose estimation](https://mlsanity.com/paper/1711.08585/) (2017)
- [Motion Guided 3D Pose Estimation from Videos](https://mlsanity.com/paper/2004.13985/) (2020)
- [Cascaded deep monocular 3D human pose estimation with evolutionary training data](https://mlsanity.com/paper/2006.07778/) (2020)
- [Compact Generalized Non-local Network](https://mlsanity.com/paper/1810.13125/) (2018)
- [Chirality Nets for Human Pose Regression](https://mlsanity.com/paper/1911.00029/) (2019)
- [Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation](https://mlsanity.com/paper/1908.08289/) (2019)

## Built on by (most cited first, 11 of 11)

- [Visual Attention Methods in Deep Learning: An In-Depth Survey](https://mlsanity.com/paper/2204.07756/) (2022)
- [HDFormer: High-order Directed Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2302.01825/) (2023)
- [3D Vision with Transformers: A Survey](https://mlsanity.com/paper/2208.04309/) (2022)
- [Occlusion Robust 3D Human Pose Estimation with StridedPoseGraphFormer and Data Augmentation](https://mlsanity.com/paper/2304.12069/) (2023)
- [SCJD: Sparse Correlation and Joint Distillation for Efficient 3D Human Pose Estimation](https://mlsanity.com/paper/2503.14097/) (2025)
- [EVOPOSE: A Recursive Transformer For 3D Human Pose Estimation With Kinematic Structure Priors](https://mlsanity.com/paper/2306.09615/) (2023)
- [Multi-hop graph transformer network for 3D human pose estimation](https://mlsanity.com/paper/2405.03055/) (2024)
- [Adaptive Convolutional Forecasting Network Based on Time Series Feature-Driven](https://mlsanity.com/paper/2405.12038/) (2024)
- [(Fusionformer):Exploiting the Joint Motion Synergy with Fusion Network Based On Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2210.04006/) (2022)
- [Refined Temporal Pyramidal Compression-and-Amplification Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2309.01365/) (2023)
- [Towards Autonomous Riding: A Review of Perception, Planning, and Control in Intelligent Two-Wheelers](https://mlsanity.com/paper/2507.11852/) (2025)

## Abstract (arXiv)

3D human pose estimation can be handled by encoding the geometric dependencies between the body parts and enforcing the kinematic constraints. Recently, Transformer has been adopted to encode the long-range dependencies between the joints in the spatial and temporal domains. While they had shown excellence in long-range dependencies, studies have noted the need for improving the locality of vision Transformers. In this direction, we propose a novel pose estimation Transformer featuring rich representations of body joints critical for capturing subtle changes across frames (i.e., inter-feature representation). Specifically, through two novel interaction modules; Cross-Joint Interaction and Cross-Frame Interaction, the model explicitly encodes the local and global dependencies between the body joints. The proposed architecture achieved state-of-the-art performance on two popular 3D human pose estimation datasets, Human3.6 and MPI-INF-3DHP. In particular, our proposed CrossFormer method boosts performance by 0.9% and 0.3%, compared to the closest counterpart, PoseFormer, using the detected 2D poses and ground-truth settings respectively.