mlsanityData to 14 Oct 2026

3D-Augmented Contrastive Knowledge Distillation for Image-based Object Pose Estimation

arXiv 2206.02531 ↗First posted 2 Jun 2022cs.CV cs.AIcited by 2

On the map: Knowledge distillation → Detection and dense prediction distillation

Abstract (arXiv)

Image-based object pose estimation sounds amazing because in real applications the shape of object is oftentimes not available or not easy to take like photos. Although it is an advantage to some extent, un-explored shape information in 3D vision learning problem looks like "flaws in jade". In this paper, we deal with the problem in a reasonable new setting, namely 3D shape is exploited in the training process, and the testing is still purely image-based. We enhance the performance of image-based methods for category-agnostic object pose estimation by exploiting 3D knowledge learned by a multi-modal method. Specifically, we propose a novel contrastive knowledge distillation framework that effectively transfers 3D-augmented image representation from a multi-modal model to an image-based model. We integrate contrastive learning into the two-stage training procedure of knowledge distillation, which formulates an advanced solution to combine these two approaches for cross-modal tasks. We experimentally report state-of-the-art results compared with existing category-agnostic image-based methods by a large margin (up to +5% improvement on ObjectNet3D dataset), demonstrating the effectiveness of our method.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation. Builds on 50 of its 69 references that are in the map. First posted on arXiv 2 Jun 2022.

Builds on

50 of 69 references.

References · 50

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  6. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  7. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  8. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  9. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  10. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,287
  11. 2016node2vec: Scalable Feature Learning for Networkscited by 1,918
  12. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,845
38 more
  1. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,380
  2. 2019Contrastive Multiview Codingcited by 1,301
  3. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,273
  4. 2019Relational Knowledge Distillationcited by 889
  5. 2019Contrastive Representation Distillationcited by 748
  6. 2019Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillationcited by 497
  7. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 495
  8. 2019A Comprehensive Overhaul of Feature Distillationcited by 348
  9. 20163D Bounding Box Estimation Using Deep Learning and Geometrycited by 278
  10. 2020Self-Supervised MultiModal Versatile Networkscited by 276
  11. 2018Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objectscited by 241
  12. 2015Unifying distillation and privileged informationcited by 233
  13. 2018Pix3D: Dataset and Methods for Single-Image 3D Shape Modelingcited by 218
  14. 2015Cross Modal Distillation for Supervision Transfercited by 175
  15. 2020Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wildcited by 142
  16. 2021Multimodal Contrastive Training for Visual Representation Learningcited by 124
  17. 2018ROI-10D: Monocular Lifting of 2D Detection to 6D Pose and Metric Shapecited by 123
  18. 2019Deep Mesh Reconstruction from Single RGB Images via Topology Modification Networkscited by 106
  19. 2019Self-supervised 6D Object Pose Estimation for Robot Manipulationcited by 101
  20. 2015Render for CNN: Viewpoint Estimation in Images Using CNNs Trained with Rendered 3D Model Viewscited by 94
  21. 20176-DoF Object Pose from Semantic Keypointscited by 89
  22. 2019LatentFusion: End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimationcited by 89
  23. 2020Self6D: Self-Supervised Monocular 6D Object Pose Estimationcited by 79
  24. 2014Viewpoints and Keypointscited by 59
  25. 2020Contrastive Distillation on Intermediate Representations for Language Model Compressioncited by 57
  26. 2021Learning Cross-modal Contrastive Features for Video Domain Adaptationcited by 47
  27. 2018StarMap for Category-Agnostic Keypoint and Viewpoint Estimationcited by 42
  28. 2019Pose from Shape: Deep Pose Estimation for Arbitrary 3D Objectscited by 39
  29. 2020On Compositions of Transformations in Contrastive Self-Supervised Learningcited by 34
  30. 2020ISD: Self-Supervised Learning by Iterative Similarity Distillationcited by 33
  31. 2021Learning an Augmented RGB Representation with Cross-Modal Knowledge Distillation for Action Detectioncited by 32
  32. 20183D Pose Estimation and 3D Model Retrieval for Objects in the Wildcited by 31
  33. 2020Self-Supervised Viewpoint Learning From Image Collectionscited by 23
  34. 2021PoseContrast: Class-Agnostic Object Viewpoint Estimation in the Wild with Pose-Aware Contrastive Learningcited by 19
  35. 2019CorNet: Generic 3D Corners for 6D Pose Estimation of New Objects without Retrainingcited by 14
  36. 2019GP2C: Geometric Projection Parameter Consensus for Joint 3D Pose and Focal Length Estimation in the Wildcited by 9
  37. 2019Few-Shot Viewpoint Estimationcited by 6
  38. 2021Can Action be Imitated? Learn to Reconstruct and Transfer Human Dynamics from Videoscited by 1

Built on by

Cited by year

Citing papers per year: 2023: 2.

2 papers.

  1. 2023PanoSwin: a Pano-style Swin Transformer for Panorama Understandingcited by 20
  2. 2023Overcoming the Trade-off Between Accuracy and Plausibility in 3D Hand Shape Reconstructioncited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2024Learning a Category-level Object Pose Estimator without Pose Annotations
  2. 2025CleanPose: Category-Level Object Pose Estimation via Causal Learning and Knowledge Distillation
  3. 2024Multi-modal Relation Distillation for Unified 3D Representation Learning
  4. 2024CLIPose: Category-Level Object Pose Estimation with Pre-trained Vision-Language Knowledge
  5. 2026Object Pose Transformer: Unifying Unseen Object Pose Estimation
  6. 2024ImageNet3D: Towards General-Purpose Object-Level 3D Understanding
  7. 2025PointDico: Contrastive 3D Representation Learning Guided by Diffusion Models
  8. 2021Online Knowledge Distillation for Efficient Pose Estimation