mlsanityData to 7 Oct 2026

Detecting Informative Channels: ActionFormer

arXiv 2505.20739 ↗First posted 27 May 2025cs.LG cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Human Activity Recognition (HAR) has recently witnessed advancements with Transformer-based models. Especially, ActionFormer shows us a new perspectives for HAR in the sense that this approach gives us additional outputs which detect the border of the activities as well as the activity labels. ActionFormer was originally proposed with its input as image/video. However, this was converted to with its input as sensor signals as well. We analyze this extensively in terms of deep learning architectures. Based on the report of high temporal dynamics which limits the model's ability to capture subtle changes effectively and of the interdependencies between the spatial and temporal features. We propose the modified ActionFormer which will decrease these defects for sensor signals. The key to our approach lies in accordance with the Sequence-and-Excitation strategy to minimize the increase in additional parameters and opt for the swish activation function to retain the information about direction in the negative range. Experiments on the WEAR dataset show that our method achieves substantial improvement of a 16.01% in terms of average mAP for inertial data.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 43 references that are in the map. First posted on arXiv 27 May 2025.

Builds on

25 of 43 references.

References · 25

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  5. 2017Squeeze-and-Excitation Networkscited by 4,763
  6. 2014Going Deeper with Convolutionscited by 4,231
  7. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  8. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  9. 2020Reformer: The Efficient Transformercited by 1,658
  10. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
  11. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  12. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
13 more
  1. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 606
  2. 2019GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyondcited by 331
  3. 2022ActionFormer: Localizing Moments of Actions with Transformerscited by 308
  4. 2023TriDet: Temporal Action Detection with Relative Boundary Modelingcited by 141
  5. 2015Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networkscited by 80
  6. 2019Probabilistic Forecasting with Temporal Convolutional Neural Networkcited by 63
  7. 2021ACTION-Net: Multipath Excitation for Action Recognitioncited by 63
  8. 2023End-to-End Temporal Action Detection with 1B Parameters Across 1000 Framescited by 47
  9. 2021Improving Deep Learning for HAR with shallow LSTMscited by 30
  10. 2023WEAR: An Outdoor Sports Dataset for Wearable and Egocentric Activity Recognitioncited by 30
  11. 2021Expansion-Squeeze-Excitation Fusion Network for Elderly Activity Recognitioncited by 23
  12. 2024Sensor Data Augmentation from Skeleton Pose Sequences for Improving Human Activity Recognitioncited by 8
  13. 2018Squeeze-and-Excitation on Spatial and Temporal Deep Feature Space for Action Recognitioncited by 1

Built on by

Not cited yet.