mlsanityData to 7 Oct 2026

Transformed ROIs for Capturing Visual Transformations in Videos

arXiv 2106.03162 ↗First posted 6 Jun 2021cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Modeling the visual changes that an action brings to a scene is critical for video understanding. Currently, CNNs process one local neighbourhood at a time, thus contextual relationships over longer ranges, while still learnable, are indirect. We present TROI, a plug-and-play module for CNNs to reason between mid-level feature representations that are otherwise separated in space and time. The module relates localized visual entities such as hands and interacting objects and transforms their corresponding regions of interest directly in the feature maps of convolutional layers. With TROI, we achieve state-of-the-art action recognition results on the large-scale datasets Something-Something-V2 and EPIC-Kitchens-100.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 41 of its 53 references that are in the map. First posted on arXiv 6 Jun 2021.

Builds on

41 of 53 references.

References · 41

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2017Mask R-CNNcited by 7,259
  5. 2016Densely Connected Convolutional Networkscited by 6,691
  6. 2016Layer Normalizationcited by 5,579
  7. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  8. 2017Non-local Neural Networkscited by 2,290
  9. 2018SlowFast Networks for Video Recognitioncited by 1,862
  10. 2021ViViT: A Video Vision Transformercited by 1,610
  11. 2017A Closer Look at Spatiotemporal Convolutions for Action Recognitioncited by 1,139
  12. 2017The "something something" video database for learning and evaluating visual common sensecited by 1,092
29 more
  1. 2016Temporal Segment Networks: Towards Good Practices for Deep Action Recognitioncited by 899
  2. 2016Simple Online and Realtime Trackingcited by 862
  3. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 852
  4. 2018Scaling Egocentric Vision: The EPIC-KITCHENS Datasetcited by 789
  5. 2019VideoBERT: A Joint Model for Video and Language Representation Learningcited by 702
  6. 2017Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classificationcited by 521
  7. 2017A simple neural network module for relational reasoningcited by 345
  8. 2019GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyondcited by 331
  9. 2017Temporal Relational Reasoning in Videoscited by 324
  10. 2018Video Action Transformer Networkcited by 316
  11. 2020Understanding Human Hands in Contact at Internet Scalecited by 281
  12. 2018Videos as Space-Time Region Graphscited by 221
  13. 2018End-to-End Dense Video Captioning with Masked Transformercited by 214
  14. 2020TDN: Temporal Difference Networks for Efficient Action Recognitioncited by 183
  15. 2020Audiovisual SlowFast Networks for Video Recognitioncited by 144
  16. 2018A2-Nets: Double Attention Networkscited by 136
  17. 2021MoViNets: Mobile Video Networks for Efficient Video Recognitioncited by 135
  18. 2018Graph-Based Global Reasoning Networkscited by 120
  19. 2019Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognitioncited by 117
  20. 2019Something-Else: Compositional Action Recognition with Spatial-Temporal Interaction Networkscited by 112
  21. 2018Actor-Centric Relation Networkcited by 87
  22. 2014Convolutional Neural Networks at Constrained Time Costcited by 81
  23. 2018Object Level Visual Reasoning in Videoscited by 67
  24. 2019More Is Less: Learning Efficient Video Representations by Big-Little Network and Depthwise Temporal Aggregationcited by 64
  25. 2017Tube Convolutional Neural Network (T-CNN) for Action Detection in Videoscited by 50
  26. 2020MVFNet: Multi-View Fusion Network for Efficient Video Recognitioncited by 36
  27. 2020Symbiotic Attention with Privileged Information for Egocentric Action Recognitioncited by 27
  28. 2021CT-Net: Channel Tensorization Network for Video Classificationcited by 26
  29. 2019LatentGNN: Learning Efficient Non-local Relations for Visual Recognitioncited by 25

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 0; 2025: 1.

3 papers.

  1. 2022Visual Object Tracking in First Person Visioncited by 25
  2. 2023Opening the Vocabulary of Egocentric Actionscited by 23
  3. 2025Memory-efficient Streaming VideoLLMs for Real-time Procedural Video Understandingcited by 18