# Deep Multi-Task Networks For Occluded Pedestrian Pose Estimation

arXiv: https://arxiv.org/abs/2206.07510 · HTML: https://mlsanity.com/paper/2206.07510/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 13 references that are in the map.
- First posted on arXiv 15 Jun 2022.

## Builds on (most cited first, 8 of 8)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Path Aggregation Network for Instance Segmentation](https://mlsanity.com/paper/1803.01534/) (2018)
- [CityPersons: A Diverse Dataset for Pedestrian Detection](https://mlsanity.com/paper/1702.05693/) (2017)
- [Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods](https://mlsanity.com/paper/2006.01423/) (2020)
- [Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd](https://mlsanity.com/paper/1807.08407/) (2018)
- [MultiPoseNet: Fast Multi-Person Pose Estimation using Pose Residual Network](https://mlsanity.com/paper/1807.04067/) (2018)
- [Spatio-Contextual Deep Network Based Multimodal Pedestrian Detection For Autonomous Driving](https://mlsanity.com/paper/2105.12713/) (2021)
- [Design of Real-time Semantic Segmentation Decoder for Automated Driving](https://mlsanity.com/paper/1901.06580/) (2019)

## Built on by (most cited first, 4 of 4)

- [Revisiting Modality Imbalance In Multimodal Pedestrian Detection](https://mlsanity.com/paper/2302.12589/) (2023)
- [BEVSeg2TP: Surround View Camera Bird's-Eye-View Based Joint Vehicle Segmentation and Ego Vehicle Trajectory Prediction](https://mlsanity.com/paper/2312.13081/) (2023)
- [SDR-GAIN: A High Real-Time Occluded Pedestrian Pose Completion Method for Autonomous Driving](https://mlsanity.com/paper/2306.03538/) (2023)
- [Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehicle](https://mlsanity.com/paper/2309.06313/) (2023)

## Abstract (arXiv)

Most of the existing works on pedestrian pose estimation do not consider estimating the pose of an occluded pedestrian, as the annotations of the occluded parts are not available in relevant automotive datasets. For example, CityPersons, a well-known dataset for pedestrian detection in automotive scenes does not provide pose annotations, whereas MS-COCO, a non-automotive dataset, contains human pose estimation. In this work, we propose a multi-task framework to extract pedestrian features through detection and instance segmentation tasks performed separately on these two distributions. Thereafter, an encoder learns pose specific features using an unsupervised instance-level domain adaptation method for the pedestrian instances from both distributions. The proposed framework has improved state-of-the-art performances of pose estimation, pedestrian detection, and instance segmentation.