BoIR: Box-Supervised Instance Representation for Multi-Person Pose Estimation
Not on the map. Why?
Abstract (arXiv)
Single-stage multi-person human pose estimation (MPPE) methods have shown great performance improvements, but existing methods fail to disentangle features by individual instances under crowded scenes. In this paper, we propose a bounding box-level instance representation learning called BoIR, which simultaneously solves instance detection, instance disentanglement, and instance-keypoint association problems. Our new instance embedding loss provides a learning signal on the entire area of the image with bounding box annotations, achieving globally consistent and disentangled instance representation. Our method exploits multi-task learning of bottom-up keypoint estimation, bounding box regression, and contrastive instance embedding learning, without additional computational cost during inference. BoIR is effective for crowded scenes, outperforming state-of-the-art on COCO val (0.8 AP), COCO test-dev (0.5 AP), CrowdPose (4.9 AP), and OCHuman (3.5 AP). Code will be available at https://github.com/uyoung-jeong/BoIR
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 37 of its 47 references that are in the map. First posted on arXiv 25 Sep 2023.
Builds on
37 of 47 references.
References · 37
- 2017Attention Is All You Needcited by 59,932
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2014Microsoft COCO: Common Objects in Contextcited by 17,498
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
- 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
- 2017Mask R-CNNcited by 7,259
- 2022A ConvNet for the 2020scited by 3,545
- 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,833
- 2018YOLOv3: An Incremental Improvementcited by 2,652
- 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
25 more
- 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
- 2019Objects as Pointscited by 1,172
- 2016Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,061
- 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
- 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
- 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 454
- 2022YOLOv6: A Single-Stage Object Detection Framework for Industrial Applicationscited by 273
- 2019HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimationcited by 247
- 2018CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmarkcited by 214
- 2016Associative Embedding: End-to-End Learning for Joint Detection and Groupingcited by 211
- 2019Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognitioncited by 135
- 2018PersonLab: Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Modelcited by 133
- 2019PifPaf: Composite Fields for Human Pose Estimationcited by 120
- 2021Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regressioncited by 112
- 2018Pose2Seg: Detection Free Human Instance Segmentationcited by 107
- 2020Rethinking the Heatmap Regression for Bottom-up Human Pose Estimationcited by 69
- 2020Differentiable Hierarchical Graph Grouping for Multi-Person Pose Estimationcited by 52
- 2023Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimationcited by 48
- 2021Multi-modal 3D Human Pose Estimation with 2D Weak Supervision in Autonomous Drivingcited by 39
- 2021FCPose: Fully Convolutional Multi-Person Pose Estimation with Dynamic Instance-Aware Convolutionscited by 34
- 2021Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimationcited by 30
- 2021Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimationcited by 20
- 2021Learning Local-Global Contextual Adaptation for Multi-Person Pose Estimationcited by 14
- 2022QueryPose: Sparse Multi-Person Pose Regression via Spatial-Aware Part-Level Querycited by 13
- 2021Sparse Object-level Supervision for Instance Segmentation with Pixel Embeddingscited by 10
Built on by
Citing papers per year: 2024: 2.
2 papers.