mlsanityData to 7 Oct 2026

Learning Discriminative Prototypes with Dynamic Time Warping

arXiv 2103.09458 ↗First posted 17 Mar 2021cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

Dynamic Time Warping (DTW) is widely used for temporal data processing. However, existing methods can neither learn the discriminative prototypes of different classes nor exploit such prototypes for further analysis. We propose Discriminative Prototype DTW (DP-DTW), a novel method to learn class-specific discriminative prototypes for temporal recognition tasks. DP-DTW shows superior performance compared to conventional DTWs on time series classification benchmarks. Combined with end-to-end deep learning, DP-DTW can handle challenging weakly supervised action segmentation problems and achieves state of the art results on standard benchmarks. Moreover, detailed reasoning on the input video is enabled by the learned action prototypes. Specifically, an action-based video summarization can be obtained by aligning the input sequence with action prototypes.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 19 of its 38 references that are in the map. First posted on arXiv 17 Mar 2021.

Builds on

19 of 38 references.

References · 19

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018The UCR Time Series Archivecited by 421
  3. 2017Soft-DTW: a Differentiable Loss Function for Time-Seriescited by 265
  4. 2017Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity-Representativeness Rewardcited by 105
  5. 2019D3TW: Discriminative Differentiable Dynamic Time Warping for Weakly Supervised Action Alignment and Segmentationcited by 83
  6. 2018Weakly-Supervised Action Segmentation with Iterative Soft Boundary Assignmentcited by 74
  7. 2016Connectionist Temporal Modeling for Weakly Supervised Action Labelingcited by 70
  8. 2017Weakly Supervised Action Learning with RNN based Fine-to-coarse Modelingcited by 70
  9. 2018Summarizing Videos with Attentioncited by 69
  10. 2018NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learningcited by 67
  11. 2019Shape and Time Distortion Loss for Training Deep Time Series Forecasting Modelscited by 66
  12. 2018Differentiable Dynamic Programming for Structured Prediction and Attentioncited by 58
7 more
  1. 2019Weakly Supervised Energy-Based Learning for Action Segmentationcited by 53
  2. 2014Weakly Supervised Action Labeling in Videos Under Ordering Constraintscited by 48
  3. 2016shapeDTW: shape Dynamic Time Warpingcited by 35
  4. 2019Temporal Transformer Networks: Joint Learning of Invariant and Discriminative Time Warpingcited by 21
  5. 2020Weakly Supervised Video Summarization by Hierarchical Reinforcement Learningcited by 18
  6. 2019Comprehensive Video Understanding: Video summarization with content-based video recommender designcited by 5
  7. 2019ILS-SUMM: Iterated Local Search for Unsupervised Video Summarizationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 8; 2023: 5; 2024: 4; 2025: 2; 2026: 2.

23 papers.

Sort:

  1. 2022Temporal Action Segmentation: An Analysis of Modern Techniquescited by 85
  2. 2022Temporal Alignment Networks for Long-term Videocited by 84
  3. 2021Drop-DTW: Aligning Common Signal Between Sequences While Dropping Outlierscited by 39
  4. 2023StepFormer: Self-supervised Step Discovery and Localization in Instructional Videoscited by 37
  5. 2022Fine-grained Temporal Contrastive Learning for Weakly-supervised Temporal Action Localizationcited by 33
  6. 2022Video-Text Representation Learning via Differentiable Weak Temporal Alignmentcited by 24
  7. 2024Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignmentcited by 13
  8. 2022Weakly-Supervised Online Action Segmentation in Multi-View Instructional Videoscited by 12
  9. 2024Causal Temporal Representation Learning with Nonstationary Sparse Transitioncited by 12
  10. 2021Hierarchical Modeling for Task Recognition and Action Segmentation in Weakly-Labeled Instructional Videoscited by 11
  11. 2023Deep Learning Approaches for Seizure Video Analysis: A Reviewcited by 6
  12. 2025CLOT: Closed Loop Optimal Transport for Unsupervised Action Segmentationcited by 6
11 more
  1. 2022Robust Time Series Dissimilarity Measure for Outlier Detection and Periodicity Detectioncited by 5
  2. 2023Deep Declarative Dynamic Time Warping for End-to-End Learning of Alignment Pathscited by 5
  3. 2022Transformers in Action: Weakly Supervised Action Segmentationcited by 4
  4. 2022Graph2Vid: Flow graph to Video Grounding for Weakly-supervised Multi-Step Localizationcited by 4
  5. 20242by2: Weakly-Supervised Learning for Global Action Segmentationcited by 3
  6. 2024Unlabeled Action Quality Assessment Based on Multi-dimensional Adaptive Constrained Dynamic Time Warpingcited by 2
  7. 2025Pose-Aware Weakly-Supervised Action Segmentationcited by 2
  8. 2023Approximating DTW with a convolutional neural network on EEG datacited by 1
  9. 2026Hierarchical Action Learning for Weakly-Supervised Action Segmentationcited by 1
  10. 2023TimewarpVAE: Simultaneous Time-Warping and Representation Learning of Trajectoriescited by 0
  11. 2026Boundary-Centric Clip-Budgeted Active Learning for Temporal Action Segmentationcited by 0