# Detecting Precise Hand Touch Moments in Egocentric Video

arXiv: https://arxiv.org/abs/2604.12343 · HTML: https://mlsanity.com/paper/2604.12343/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 44 references that are in the map.
- First posted on arXiv 14 Apr 2026.

## Builds on (most cited first, 30 of 37)

- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Grounded Language-Image Pre-training](https://mlsanity.com/paper/2112.03857/) (2021)
- [Designing Network Design Spaces](https://mlsanity.com/paper/2003.13678/) (2020)
- [Soft-NMS -- Improving Object Detection With One Line of Code](https://mlsanity.com/paper/1704.04503/) (2017)
- [ResNet strikes back: An improved training procedure in timm](https://mlsanity.com/paper/2110.00476/) (2021)
- [ActionFormer: Localizing Moments of Actions with Transformers](https://mlsanity.com/paper/2202.07925/) (2022)
- [Video Classification with Channel-Separated Convolutional Networks](https://mlsanity.com/paper/1904.02811/) (2019)
- [Understanding Human Hands in Contact at Internet Scale](https://mlsanity.com/paper/2006.06669/) (2020)
- [GLIPv2: Unifying Localization and Vision-Language Understanding](https://mlsanity.com/paper/2206.05836/) (2022)
- [Video Transformer Network](https://mlsanity.com/paper/2102.00719/) (2021)
- [G-TAD: Sub-Graph Localization for Temporal Action Detection](https://mlsanity.com/paper/1911.11462/) (2019)
- [ASFormer: Transformer for Action Segmentation](https://mlsanity.com/paper/2110.08568/) (2021)
- [Temporal Pyramid Network for Action Recognition](https://mlsanity.com/paper/2004.03548/) (2020)
- [SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videos](https://mlsanity.com/paper/1804.04527/) (2018)
- [TriDet: Temporal Action Detection with Relative Boundary Modeling](https://mlsanity.com/paper/2303.07347/) (2023)
- [EPIC-KITCHENS VISOR Benchmark: VIdeo Segmentations and Object Relations](https://mlsanity.com/paper/2209.13064/) (2022)
- [SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos](https://mlsanity.com/paper/2011.13367/) (2020)
- [Gate-Shift Networks for Video Action Recognition](https://mlsanity.com/paper/1912.00381/) (2019)
- [Human Hands as Probes for Interactive Object Understanding](https://mlsanity.com/paper/2112.09120/) (2021)
- [Detecting Human-Object Contact in Images](https://mlsanity.com/paper/2303.03373/) (2023)
- [Spotting Temporally Precise, Fine-Grained Events in Video](https://mlsanity.com/paper/2207.10213/) (2022)
- [Detecting Hands and Recognizing Physical Contact in the Wild](https://mlsanity.com/paper/2010.09676/) (2020)
- [3D Hand Pose Estimation in Everyday Egocentric Images](https://mlsanity.com/paper/2312.06583/) (2023)
- [Feature Combination Meets Attention: Baidu Soccer Embeddings and Transformer based Temporal Detection](https://mlsanity.com/paper/2106.14447/) (2021)
- [T-DEED: Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in Sports Videos](https://mlsanity.com/paper/2404.05392/) (2024)
- [GTA: Global Temporal Attention for Video Action Understanding](https://mlsanity.com/paper/2012.08510/) (2020)
- [COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformers](https://mlsanity.com/paper/2309.01270/) (2023)
- [Contextual Attention for Hand Detection in the Wild](https://mlsanity.com/paper/1904.04882/) (2019)

## Built on by (most cited first, 2 of 2)

- [Iterative Erasure Count Is Not an Affine-Invariant Concept Dimension](https://mlsanity.com/paper/2608.10566/) (2026)
- [Empirical Evaluation of Multi-Modal Touch Detection in Over-the-Shoulder Video Surveillance](https://mlsanity.com/paper/2606.29504/) (2026)

## Abstract (arXiv)

We address the challenging task of detecting the precise moment when hands make contact with objects in egocentric videos. This frame-level detection is crucial for augmented reality, human-computer interaction, assistive technologies, and robot learning applications, where contact onset signals action initiation or completion. Temporally precise detection is particularly challenging due to subtle hand motion variations near contact, frequent occlusions, fine-grained manipulation patterns, and the inherent motion dynamics of first-person perspectives. To tackle these challenges, we propose a Hand-informed Context Enhanced module (HiCE; pronounced `high-see') that leverages spatiotemporal features from hand regions and their surrounding context through cross-attention mechanisms, learning to identify potential contact patterns. Our approach is further refined with a grasp-aware loss and soft label that emphasizes hand pose patterns and movement dynamics characteristic of touch events, enabling the model to distinguish between near-contact and actual contact frames. We also introduce TouchMoment, an egocentric dataset containing 4,021 videos and 8,456 annotated contact moments spanning over one million frames. Experiments on TouchMoment show that, under a strict evaluation criterion that counts a prediction as correct only if it falls within a two-frame tolerance of the ground-truth moment, our method achieves substantial gains and outperforms state-of-the-art event-spotting baselines by 16.91% average precision.