mlsanityData to 14 Oct 2026

MobileVOS: Real-Time Video Object Segmentation Contrastive Learning meets Knowledge Distillation

arXiv 2303.07815 ↗First posted 14 Mar 2023cs.CVcited by 16

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

Abstract (arXiv)

This paper tackles the problem of semi-supervised video object segmentation on resource-constrained devices, such as mobile phones. We formulate this problem as a distillation task, whereby we demonstrate that small space-time-memory networks with finite memory can achieve competitive results with state of the art, but at a fraction of the computational cost (32 milliseconds per frame on a Samsung Galaxy S22). Specifically, we provide a theoretically grounded framework that unifies knowledge distillation with supervised contrastive representation learning. These models are able to jointly benefit from both pixel-wise contrastive learning and distillation from a pre-trained teacher. We validate this loss by achieving competitive J&F to state of the art on both the standard DAVIS and YouTube benchmarks, despite running up to 5x faster, and with 32x fewer parameters.

Key facts. Cited by 16 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 64 of its 79 references that are in the map. First posted on arXiv 14 Mar 2023.

Builds on

64 of 79 references.

References · 64

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  5. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  6. 2020Supervised Contrastive Learningcited by 2,829
  7. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  8. 2014FitNets: Hints for Thin Deep Netscited by 1,653
  9. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,396
  10. 2016Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfercited by 1,077
  11. 2017The 2017 DAVIS Challenge on Video Object Segmentationcited by 957
  12. 2019Contrastive Representation Distillationcited by 748
38 more
  1. 2018Born Again Neural Networkscited by 546
  2. 2022XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Modelcited by 439
  3. 2019Video Object Segmentation using Space-Time Memory Networkscited by 438
  4. 2018YouTube-VOS: A Large-Scale Video Object Segmentation Benchmarkcited by 383
  5. 2021Distilling Knowledge via Knowledge Reviewcited by 288
  6. 2022Unsupervised Semantic Segmentation by Distilling Feature Correspondencescited by 241
  7. 2016One-Shot Video Object Segmentationcited by 235
  8. 2021Associating Objects with Transformers for Video Object Segmentationcited by 226
  9. 2021Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentationcited by 203
  10. 2019FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentationcited by 178
  11. 2020Channel-wise Knowledge Distillation for Dense Predictioncited by 163
  12. 2016Learning Video Object Segmentation from Static Imagescited by 151
  13. 2021Self-supervised Video Object Segmentation by Motion Groupingcited by 147
  14. 2020Collaborative Video Object Segmentation by Foreground-Background Integrationcited by 142
  15. 2020Knowledge Distillation Meets Self-Supervisioncited by 142
  16. 2021A Survey on Deep Learning Technique for Video Segmentationcited by 132
  17. 2021Modular Interactive Video Object Segmentation: Interaction-to-Mask, Propagation and Difference-Aware Fusioncited by 129
  18. 2020Video Object Segmentation with Episodic Graph Memory Networkscited by 118
  19. 2020Kernelized Memory Network for Video Object Segmentationcited by 111
  20. 2022On the duality between contrastive and non-contrastive self-supervised learningcited by 108
  21. 2021SSTVOS: Sparse Spatiotemporal Transformers for Video Object Segmentationcited by 106
  22. 2020Video Object Segmentation with Adaptive Feature Bank and Uncertain-Region Refinementcited by 101
  23. 2018Efficient Video Object Segmentation via Network Modulationcited by 98
  24. 2021FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpaintingcited by 95
  25. 2017SegFlow: Joint Learning for Video Object Segmentation and Optical Flowcited by 92
  26. 2018VideoMatch: Matching based Video Object Segmentationcited by 90
  27. 2021Efficient Regional Memory Network for Video Object Segmentationcited by 86
  28. 2018PReMVOS: Proposal-generation, Refinement and Merging for Video Object Segmentationcited by 81
  29. 2018Benchmark Analysis of Representative Deep Neural Network Architecturescited by 77
  30. 2018Blazingly Fast Video Object Segmentation with Pixel-Wise Metric Learningcited by 74
  31. 2021SwiftNet: Real-time Video Object Segmentationcited by 67
  32. 2020Fast Video Object Segmentation using the Global Context Modulecited by 65
  33. 2020Wasserstein Contrastive Representation Distillationcited by 64
  34. 2022PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functionscited by 63
  35. 2021Joint Inductive and Transductive Learning for Video Object Segmentationcited by 61
  36. 2018Video Object Segmentation with Joint Re-identification and Attention-Aware Mask Propagationcited by 56
  37. 2022Recurrent Dynamic Embedding for Video Object Segmentationcited by 52
  38. 2018Dynamic Video Segmentation Networkcited by 51

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 9; 2025: 3; 2026: 2.

16 papers.

Sort:

By area:

  1. 2023Putting the Object Back into Video Object Segmentationcited by 188
  2. 2024VkD: Improving Knowledge Distillation using Orthogonal Projectionscited by 26
  3. 2024DaCapo: Accelerating Continuous Learning in Autonomous Systems for Video Analyticscited by 8
  4. 2024CKD: Contrastive Knowledge Distillation from A Sample-wise Perspectivecited by 7
  5. 2024LiVOS: Light Video Object Segmentation with Gated Linear Matchingcited by 7
  6. 2024Efficient Video Object Segmentation via Modulated Cross-Attention Memorycited by 3
  7. 2024DiDA: Video Object Segmentation with Distillation Learning of Deformable Attentioncited by 2
  8. 2025InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspectivecited by 1
  9. 2023TrickVOS: A Bag of Tricks for Video Object Segmentationcited by 0
  10. 2024Learning to Project for Cross-Task Knowledge Distillationcited by 0
  11. 2024Dynamic in Static: Hybrid Visual Correspondence for Self-Supervised Video Object Segmentationcited by 0
  12. 2024ActivityCLIP: Enhancing Group Activity Recognition by Mining Complementary Information from Text to Supplement Image Modalitycited by 0
4 more
  1. 2025A Survey on Efficiency Optimization Techniques for DNN-based Video Analytics: Process Systems, Algorithms, and Applicationscited by 0
  2. 2025PPMStereo: Pick-and-Play Memory Construction for Consistent Dynamic Stereo Matchingcited by 0
  3. 2026MobileSAM2: Lightweight Segment Anything for Spatial Intelligencecited by 0
  4. 2026SWT: Self-Supervised Video Object Segmentation via Sliding, Wavelet and Transportationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Augmentation-Free Dense Contrastive Knowledge Distillation for Efficient Semantic Segmentation
  2. 2025CAST: Contrastive Adaptation and Distillation for Semi-Supervised Instance Segmentation
  3. 2026Report of the 8th LSVOS Challenge: Complex and Multimodal Video Object Segmentation
  4. 2022Delta Distillation for Efficient Video Processing
  5. 2026Training a Student Expert via Semi-Supervised Foundation Model Distillation
  6. 2026Bootstrapping Video Semantic Segmentation Model via Distillation-assisted Test-Time Adaptation
  7. 2026SWARD: Stochastic Window-Attention-Based Relational Distillation for Cross-Architectural Semantic Segmentation
  8. 2025UINO-FSS: Unifying Representation Learning and Few-shot Segmentation via Hierarchical Distillation and Mamba-HyperCorrelation