mlsanityData to 7 Oct 2026

Knowledge Distillation for 6D Pose Estimation by Aligning Distributions of Local Predictions

arXiv 2205.14971 ↗First posted 30 May 2022cs.CV cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Knowledge distillation facilitates the training of a compact student network by using a deep teacher one. While this has achieved great success in many tasks, it remains completely unstudied for image-based 6D object pose estimation. In this work, we introduce the first knowledge distillation method driven by the 6D pose estimation task. To this end, we observe that most modern 6D pose estimation frameworks output local predictions, such as sparse 2D keypoints or dense representations, and that the compact student network typically struggles to predict such local quantities precisely. Therefore, instead of imposing prediction-to-prediction supervision from the teacher to the student, we propose to distill the teacher's distribution of local predictions into the student network, facilitating its training. Our experiments on several benchmarks show that our distillation method yields state-of-the-art results with different compact student models and for both keypoint-based and dense prediction-based architectures.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 33 of its 56 references that are in the map. First posted on arXiv 30 May 2022.

Builds on

33 of 56 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  3. 2017Mask R-CNNcited by 7,259
  4. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  5. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  6. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  7. 2018YOLOv3: An Incremental Improvementcited by 2,652
  8. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,277
  9. 2014FitNets: Hints for Thin Deep Netscited by 1,619
  10. 2016Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfercited by 1,059
  11. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
  12. 2019Contrastive Representation Distillationcited by 734
21 more
  1. 2015Optimal Transport for Domain Adaptationcited by 457
  2. 2018PVNet: Pixel-wise Voting Network for 6DoF Pose Estimationcited by 385
  3. 2019A Comprehensive Overhaul of Feature Distillationcited by 341
  4. 2019Structured Knowledge Distillation for Dense Predictioncited by 277
  5. 2017SSD-6D: Making RGB-based 3D detection and 6D pose estimation great againcited by 243
  6. 2019Model Fusion via Optimal Transportcited by 217
  7. 2020BOP Challenge 2020 on 6D Object Localizationcited by 192
  8. 2019Distilling Object Detectors with Fine-grained Feature Imitationcited by 188
  9. 2021GDRNPP: A Geometry-guided and Fully Learning-based Object Pose Estimatorcited by 148
  10. 2022ZebraPose: Coarse to Fine Surface Encoding for 6DoF Object Pose Estimationcited by 115
  11. 2019Knowledge Adaptation for Efficient Semantic Segmentationcited by 105
  12. 2021Distilling Object Detectors via Decoupled Featurescited by 105
  13. 2018Segmentation-driven 6D Object Pose Estimationcited by 104
  14. 2019Single-Stage 6D Object Pose Estimationcited by 88
  15. 2021SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimationcited by 88
  16. 2018Making Deep Heatmaps Robust to Partial Occlusions for 3D Object Pose Estimationcited by 74
  17. 2021Wide-Depth-Range 6D Object Pose Estimation in Spacecited by 45
  18. 2019Progressive-X: Efficient, Anytime, Multi-Model Fitting Algorithmcited by 43
  19. 2017An Efficient Algebraic Solution to the Perspective-Three-Point Problemcited by 29
  20. 2022HRPose: Real-Time High-Resolution 6D Pose Estimation Network Using Knowledge Distillationcited by 11
  21. 2021Distilling Image Classifiers in Object Detectorscited by 5

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 2; 2025: 4.

9 papers.

  1. 2024Deep Learning-Based Object Pose Estimation: A Comprehensive Surveycited by 39
  2. 20236D-Diff: A Keypoint Diffusion Framework for 6D Object Pose Estimationcited by 24
  3. 2023Challenges for Monocular 6D Object Pose Estimation in Roboticscited by 21
  4. 2024MRC-Net: 6-DoF Pose Estimation with MultiScale Residual Correlationcited by 11
  5. 2025Uncertainty-Aware Knowledge Distillation for Compact and Efficient 6DoF Pose Estimationcited by 5
  6. 2023Pseudo-keypoint RKHS Learning for Self-supervised 6DoF Pose Estimationcited by 4
  7. 2025Distilling 3D distinctive local descriptors for 6D pose estimationcited by 4
  8. 2025FPG-NAS: FLOPs-Aware Gated Differentiable Neural Architecture Search for Efficient 6DoF Pose Estimationcited by 1
  9. 2025PoseIDON: 6DoF Pose Estimation with Foundation Model Features for Marine Sediment Burial Mappingcited by 0