Papers
Topics
Authors
Recent
Search
2000 character limit reached

Camera clustering for scalable stream-based active distillation

Published 16 Apr 2024 in cs.CV | (2404.10411v1)

Abstract: We present a scalable framework designed to craft efficient lightweight models for video object detection utilizing self-training and knowledge distillation techniques. We scrutinize methodologies for the ideal selection of training images from video streams and the efficacy of model sharing across numerous cameras. By advocating for a camera clustering methodology, we aim to diminish the requisite number of models for training while augmenting the distillation dataset. The findings affirm that proper camera clustering notably amplifies the accuracy of distilled models, eclipsing the methodologies that employ distinct models for each camera or a universal model trained on the aggregate camera data.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (54)
  1. Knowledge distillation: A good teacher is patient and consistent. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10925–10934, June 2022.
  2. Stream-based active distillation for scalable model deployment. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pages 4998–5006, June 2023.
  3. Mark D. Hill. What is scalability? SIGARCH Comput. Archit. News, 18(4):18–21, dec 1990. ISSN 0163-5964.
  4. Comparing the scalability of communication networks and systems. IEEE Access, pages 1–1, 2023. doi:10.1109/ACCESS.2023.3314201.
  5. Cluster-based scalable network services. In Proceedings of the Sixteenth ACM Symposium on Operating Systems Principles, SOSP ’97, page 78–91, New York, NY, USA, 1997. Association for Computing Machinery. ISBN 0897919165.
  6. Distilling the knowledge in a neural network. NIPS 2014 Deep Learning Workshop, 2015.
  7. Model compression and acceleration for deep neural networks: The principles, progress, and challenges. IEEE Signal Processing Magazine, 2018.
  8. Designing and training of lightweight neural networks on edge devices using early halting in knowledge distillation. IEEE Transactions on Mobile Computing, 2023.
  9. Federated learning by employing knowledge distillation on edge devices with limited hardware resources. Neurocomputing, 531:87–99, 2023.
  10. Semi-supervised learning for domain adaptation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, 2019.
  11. Arthus: Adaptive real-time human segmentation in sports through online distillation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pages 0–0, 2019.
  12. Online model distillation for efficient video inference. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3573–3582, 2019.
  13. Towards automatic model specialization for edge video analytics. Future Generation Computer Systems, 2022.
  14. Delta distillation for efficient video processing. In Shai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, and Tal Hassner, editors, Computer Vision – ECCV 2022, pages 213–229, Cham, 2022. Springer Nature Switzerland. ISBN 978-3-031-19833-5.
  15. Group distributionally robust knowledge distillation. In International Workshop on Machine Learning in Medical Imaging, pages 234–242. Springer, 2023.
  16. Robust active distillation. arXiv preprint arXiv:2210.01213, 2022.
  17. Adversarially robust distillation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 3996–4003, 2020.
  18. Burr Settles. Active learning literature survey. Computer Sciences Technical Report, University of Wisconsin–Madison, 2009.
  19. Active learning for convolutional neural networks: A core-set approach. In ICLR 2018, 8 2017.
  20. Learning loss for active learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2019.
  21. Deep batch active learning by diverse, uncertain gradient lower bounds. In 2020 International Conference on Learning Representations, 6 2019.
  22. A convex optimization framework for active learning. In Proceedings of the IEEE International Conference on Computer Vision, pages 209–216. Institute of Electrical and Electronics Engineers Inc., 2013. ISBN 9781479928392. doi:10.1109/ICCV.2013.33.
  23. Contextual diversity for active learning. In European Conference on Computer Vision (ECCV) 2020, 8 2020.
  24. Active domain adaptation via clustering uncertainty-weighted embeddings. In International Conference on Computer Vision (ICCV) 2021, 2020.
  25. Variational adversarial active learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 5972–5981, 2019.
  26. Multiple instance active learning for object detection. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 4 2021. URL http://arxiv.org/abs/2104.02324.
  27. Stream-based active learning with linear models. Knowledge-Based Systems, 254:109664, 10 2022a. ISSN 09507051. doi:10.1016/j.knosys.2022.109664.
  28. Robust online active learning. Quality and Reliability Engineering International, ENBIS Special Issue, 2023. doi:https://doi.org/10.1002/qre.3392. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/qre.3392.
  29. Online active learning for soft sensor development using semi-supervised autoencoders. In ICML 2022 Workshop on Adaptive Experimental Design and Active Learning in the Real World, 2022b. doi:10.48550/arXiv.2212.13067. URL https://arxiv.org/abs/2212.13067.
  30. Streaming machine learning and online active learning for automated visual inspection. IFAC-PapersOnLine, 55(2):277–282, 2022. ISSN 2405-8963. doi:https://doi.org/10.1016/j.ifacol.2022.04.206. URL https://www.sciencedirect.com/science/article/pii/S2405896322002075. 14th IFAC Workshop on Intelligent Manufacturing Systems IMS 2022.
  31. Stream-based active learning for efficient and adaptive classification of 3d objects. In Proceedings - IEEE International Conference on Robotics and Automation, volume 2016-June, pages 227–233. Institute of Electrical and Electronics Engineers Inc., 6 2016. ISBN 9781467380263. doi:10.1109/ICRA.2016.7487138.
  32. Video summaries through mosaic-based shot and scene clustering. In Computer Vision—ECCV 2002: 7th European Conference on Computer Vision Copenhagen, Denmark, May 28–31, 2002 Proceedings, Part IV 7, pages 388–402. Springer, 2002.
  33. Hierarchical semantic contrast for scene-aware video anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22846–22856, 2023.
  34. An intelligent recommendation system model based on style for virtual home furnishing in three-dimensional scene. In 2013 International Symposium on Computational and Business Intelligence, pages 213–216. IEEE, 2013.
  35. Hierarchical clustering and refinement for generalized multi-camera person tracking. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5519–5528, 2023.
  36. Multi-camera trajectory matching based on hierarchical clustering and constraints. Multimedia Tools and Applications, pages 1–24, 2023.
  37. Correspondence-free multi-camera activity analysis and scene modeling. In 2008 IEEE Conference on Computer Vision and Pattern Recognition, pages 1–8. IEEE, 2008.
  38. An occlusion-aware multi-target multi-camera tracking system. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4173–4182, 2021.
  39. Multi-camera video scene graphs for surveillance videos indexing and retrieval. In 2021 IEEE International Conference on Image Processing (ICIP), pages 2383–2387. IEEE, 2021.
  40. Visual event recognition using decision trees. Multimedia Tools and Applications, 50:95–121, 2010.
  41. Multiview clustering via hypergraph induced semi-supervised symmetric nonnegative matrix factorization. IEEE Transactions on Circuits and Systems for Video Technology, 33(10):5510–5524, 2023. doi:10.1109/TCSVT.2023.3258926.
  42. Fast multi-view clustering via ensembles: Towards scalability, superiority, and simplicity. IEEE Transactions on Knowledge and Data Engineering, 35(11):11388–11402, 2023. doi:10.1109/TKDE.2023.3236698.
  43. Pseudo-labeling and confirmation bias in deep semi-supervised learning. In 2020 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE, 2020.
  44. Ccsd: cross-camera self-distillation for unsupervised person re-identification. Visual Intelligence, 1(1):27, 2023.
  45. Model compression. In Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’06, page 535–541, New York, NY, USA, 2006. Association for Computing Machinery. ISBN 1595933395.
  46. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 779–788, 2016.
  47. R. Sibson. SLINK: An optimally efficient algorithm for the single-link cluster method. The Computer Journal, 16(1):30–34, 01 1973. ISSN 0010-4620.
  48. Walt: Watch and learn 2d amodal representation from time-lapse imagery. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 9356–9366, June 2022.
  49. Microsoft COCO: common objects in context. CoRR, abs/1405.0312, 2014.
  50. Ultralytics YOLO, January 2023. URL https://github.com/ultralytics/ultralytics.
  51. Robert M French. Catastrophic forgetting in connectionist networks. Trends in cognitive sciences, 3(4):128–135, 1999.
  52. Meta-learning-based incremental few-shot object detection. IEEE Transactions on Circuits and Systems for Video Technology, 32(4):2158–2169, 2022. doi:10.1109/TCSVT.2021.3088545.
  53. Semantic knowledge guided class-incremental learning. IEEE Transactions on Circuits and Systems for Video Technology, 33(10):5921–5931, 2023. doi:10.1109/TCSVT.2023.3262739.
  54. Robust aggregation for federated learning. IEEE Transactions on Signal Processing, 70:1142–1154, 2022. doi:10.1109/TSP.2022.3153135.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up