OmViD: Omni-supervised active learning for video action detection
Not on the map. Why?
Abstract (arXiv)
Video action detection requires dense spatio-temporal annotations, which are both challenging and expensive to obtain. However, real-world videos often vary in difficulty and may not require the same level of annotation. This paper analyzes the appropriate annotation types for each sample and their impact on spatio-temporal video action detection. It focuses on two key aspects: 1) how to obtain varying levels of annotation for videos, and 2) how to learn action detection from different annotation types. The study explores video-level tags, points, scribbles, bounding boxes, and pixel-level masks. First, a simple active learning strategy is proposed to estimate the necessary annotation type for each video. Then, a novel spatio-temporal 3D-superpixel approach is introduced to generate pseudo-labels from these annotations, enabling effective training. The approach is validated on UCF101-24 and JHMDB-21 datasets, significantly cutting annotation costs with minimal performance loss.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 54 of its 82 references that are in the map. First posted on arXiv 19 Aug 2025.
Builds on
54 of 82 references.
References · 54
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Microsoft COCO: Common Objects in Contextcited by 17,608
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
- 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
- 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
- 2015Fast R-CNNcited by 2,933
- 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
- 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
- 2016YOLO9000: Better, Faster, Strongercited by 1,431
- 2019MixMatch: A Holistic Approach to Semi-Supervised Learningcited by 1,391
- 2017Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?cited by 551
38 more
- 2017AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actionscited by 449
- 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 432
- 2019Video Instance Segmentationcited by 393
- 2016ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentationcited by 331
- 2021End-to-End Semi-Supervised Object Detection with Soft Teachercited by 316
- 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
- 2016Simple Does It: Weakly Supervised Instance and Semantic Segmentationcited by 227
- 2017Cost-Effective Active Learning for Deep Image Classificationcited by 166
- 2017Learning Active Learning from Datacited by 101
- 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
- 2017Action Tubelet Detector for Spatio-Temporal Action Localizationcited by 77
- 2021Multiple instance active learning for object detectioncited by 77
- 2020Superpixel Segmentation with Fully Convolutional Networkscited by 72
- 2020Asynchronous Interaction Aggregation for Action Detectioncited by 65
- 2019Active Learning for Deep Detection Neural Networkscited by 62
- 2021Points as Queries: Weakly Semi-supervised Object Detection by Pointscited by 60
- 2016Online Real-time Multiple Spatiotemporal Action Localisation and Predictioncited by 57
- 2017Cross-Domain Self-supervised Multi-task Feature Learning using Synthetic Imagerycited by 56
- 2020Actions as Moving Pointscited by 55
- 2020Not All Unlabeled Data are Equal: Learning to Weight Data in Semi-supervised Learningcited by 53
- 2019STEP: Spatio-Temporal Progressive Learning for Video Action Detectioncited by 51
- 2017Tube Convolutional Neural Network (T-CNN) for Action Detection in Videoscited by 50
- 2018A Structured Model For Action Detectioncited by 43
- 2020Improving Object Detection with Selective Self-supervised Self-trainingcited by 35
- 2018VideoCapsuleNet: A Simplified Network for Action Detectioncited by 30
- 2022Omni-DETR: Omni-Supervised Object Detection with Transformerscited by 29
- 2017Revisiting knowledge transfer for training object class detectorscited by 26
- 2020UFO2: A Unified Framework towards Omni-supervised Object Detectioncited by 26
- 2019Temporal Coherence for Active Learning in Videoscited by 24
- 2022End-to-End Semi-Supervised Learning for Video Action Detectioncited by 17
- 2022Active Learning Strategies for Weakly-supervised Object Detectioncited by 17
- 2018A flexible model for training action localization with varying levels of supervisioncited by 16
- 2020Uncertainty-Aware Weakly Supervised Action Detection from Untrimmed Videoscited by 15
- 2018Pointly-Supervised Action Localizationcited by 12
- 2022Video Action Detection: Analysing Limitations and Challengescited by 12
- 2025STPro: Spatial and Temporal Progressive Learning for Weakly Supervised Spatio-Temporal Groundingcited by 11
- 2025Contextual Self-paced Learning for Weakly Supervised Spatio-Temporal Video Groundingcited by 10
- 2019Activity Driven Weakly Supervised Object Detectioncited by 9
50 most cited of 54 All 54 (JSON)
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.