mlsanityData to 14 Oct 2026

Localization with Sampling-Argmax

arXiv 2110.08825 ↗First posted 17 Oct 2021cs.CV cs.LGcited by 11

Not on the map. Why?

Abstract (arXiv)

Soft-argmax operation is commonly adopted in detection-based methods to localize the target position in a differentiable manner. However, training the neural network with soft-argmax makes the shape of the probability map unconstrained. Consequently, the model lacks pixel-wise supervision through the map during training, leading to performance degradation. In this work, we propose sampling-argmax, a differentiable training method that imposes implicit constraints to the shape of the probability map by minimizing the expectation of the localization error. To approximate the expectation, we introduce a continuous formulation of the output distribution and develop a differentiable sampling process. The expectation can be approximated by calculating the average error of all samples drawn from the output distribution. We show that sampling-argmax can seamlessly replace the conventional soft-argmax operation on various localization tasks. Comprehensive experiments demonstrate the effectiveness and flexibility of the proposed method. Code is available at https://github.com/Jeff-sjtu/sampling-argmax

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 33 of its 57 references that are in the map. First posted on arXiv 17 Oct 2021.

Builds on

33 of 57 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  4. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  5. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  6. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  7. 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
  8. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 360
  9. 2020HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimationcited by 307
  10. 2017Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperationcited by 289
  11. 2017Integral Human Pose Regressioncited by 256
  12. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
21 more
  1. 2018CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmarkcited by 214
  2. 2015Deep Spatial Autoencoders for Visuomotor Learningcited by 153
  3. 2014A* Samplingcited by 151
  4. 2019Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Imagecited by 132
  5. 2021Human Pose Regression with Residual Log-likelihood Estimationcited by 131
  6. 2019DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatchcited by 114
  7. 20182D/3D Pose Estimation and Action Recognition using Multitask Deep Learningcited by 103
  8. 2018Numerical Coordinate Regression with Convolutional Neural Networkscited by 72
  9. 2016SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentationcited by 70
  10. 2019SFNet: Learning Object-aware Semantic Correspondencecited by 60
  11. 2017Human Pose Regression by Combining Indirect Part Detection and Contextual Informationcited by 52
  12. 2020KeypointNet: A Large-scale 3D Keypoint Dataset Aggregated from Numerous Human Annotationscited by 51
  13. 2018PoseFix: Model-agnostic General Human Pose Refinement Networkcited by 47
  14. 2020HMOR: Hierarchical Multi-Person Ordinal Relations for Monocular Multi-Person 3D Pose Estimationcited by 44
  15. 2017Improving Landmark Localization with Semi-Supervised Learningcited by 39
  16. 2021Skeleton Merger: an Unsupervised Aligned Keypoint Detectorcited by 27
  17. 2020Learning to Orient Surfaces by Self-supervised Spherical CNNscited by 26
  18. 2015Learning to Linearize Under Uncertaintycited by 25
  19. 2016Semantic Parsing with Semi-Supervised Sequential Autoencoderscited by 11
  20. 2019Grand Challenge of 106-Point Facial Landmark Localizationcited by 7
  21. 2020Cylindrical Convolutional Networks for Joint Object Detection and Viewpoint Estimationcited by 7

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 4; 2025: 3.

11 papers.

  1. 2022D&D: Learning Human Dynamics from Dynamic Cameracited by 41
  2. 2024LocLLM: Exploiting Generalizable Human Keypoint Localization via Large Language Modelcited by 17
  3. 2024Deep Reinforcement Learning for Time-Critical Wilderness Search And Rescue Using Dronescited by 5
  4. 2023TIC-TAC: A Framework for Improved Covariance Estimation in Deep Heteroscedastic Regressioncited by 4
  5. 2024Pixel-Wise Symbol Spotting via Progressive Points Location for Parsing CAD Imagescited by 2
  6. 2025Towards Self-Supervised Covariance Estimation in Deep Heteroscedastic Regressioncited by 2
  7. 2024The Sampling-Gaussian for stereo matchingcited by 1
  8. 2023CornerFormer: Boosting Corner Representation for Fine-Grained Structured Reconstructioncited by 0
  9. 2023Topology-inspired Cross-domain Network for Developmental Cervical Stenosis Quantificationcited by 0
  10. 2025Incremental Object Keypoint Learningcited by 0
  11. 2025Diagnose Like A REAL Pathologist: An Uncertainty-Focused Approach for Trustworthy Multi-Resolution Multiple Instance Learningcited by 0