# Beyond Pairwise Relations: Dynamic Manipulation Hypergraphs for Vision-Based Human Activity Recognition

arXiv: https://arxiv.org/abs/2607.14350 · HTML: https://mlsanity.com/paper/2607.14350/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 29 references that are in the map.
- First posted on arXiv 15 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [Rescaling Egocentric Vision](https://mlsanity.com/paper/2006.13256/) (2020)
- [VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://mlsanity.com/paper/2303.16727/) (2023)
- [Hypergraph Neural Networks](https://mlsanity.com/paper/1809.09401/) (2018)
- [HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction](https://mlsanity.com/paper/2203.01577/) (2022)
- [InternVideo2: Scaling Foundation Models for Multimodal Video Understanding](https://mlsanity.com/paper/2403.15377/) (2024)
- [ARCTIC: A Dataset for Dexterous Bimanual Hand-Object Manipulation](https://mlsanity.com/paper/2204.13662/) (2022)
- [Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activities](https://mlsanity.com/paper/2203.14712/) (2022)
- [H2O: Two Hands Manipulating Objects for First Person Interaction Recognition](https://mlsanity.com/paper/2104.11181/) (2021)
- [EPIC-KITCHENS VISOR Benchmark: VIdeo Segmentations and Object Relations](https://mlsanity.com/paper/2209.13064/) (2022)
- [HD-EPIC: A Highly-Detailed Egocentric Video Dataset](https://mlsanity.com/paper/2502.04144/) (2025)
- [Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and Applications](https://mlsanity.com/paper/2208.03826/) (2022)
- [Action Scene Graphs for Long-Form Understanding of Egocentric Videos](https://mlsanity.com/paper/2312.03391/) (2023)
- [Forecasting Action through Contact Representations from First Person Video](https://mlsanity.com/paper/2102.00649/) (2021)
- [3Mformer: Multi-order Multi-mode Transformer for Skeletal Action Recognition](https://mlsanity.com/paper/2303.14474/) (2023)
- [Hypergraph Transformer for Skeleton-based Action Recognition](https://mlsanity.com/paper/2211.09590/) (2022)
- [HyperGLM: HyperGraph for Video Scene Graph Generation and Anticipation](https://mlsanity.com/paper/2411.18042/) (2024)
- [A Hierarchical Graph-based Approach for Recognition and Description Generation of Bimanual Actions in Videos](https://mlsanity.com/paper/2310.00670/) (2023)
- [Multi Sentence Description of Complex Manipulation Action Videos](https://mlsanity.com/paper/2311.07285/) (2023)
- [EfficientGFormer: Multimodal Brain Tumor Segmentation via Pruned Graph-Augmented Transformer](https://mlsanity.com/paper/2508.01465/) (2025)
- [Autoregressive Adaptive Hypergraph Transformer for Skeleton-based Activity Recognition](https://mlsanity.com/paper/2411.05692/) (2024)

## Abstract (arXiv)

Fine-grained manipulation recognition requires modeling evolving relations among hands, objects, tools, and supporting surfaces. Conventional graph-based methods use pairwise edges that can fragment a coordinated event into disconnected binary relations. We propose a dynamic manipulation hypergraph framework that represents multi-entity configurations as higher-order relational units. At each temporal step, relevant entities are encoded using appearance, spatial, motion, and semantic-role features. Hyperedge candidates are instantiated and ranked using proximity, contact, and motion-coupling predicates. A hypergraph reasoning network performs node-to-hyperedge and hyperedge-to-node message passing, followed by temporal attention over the evolving interaction structure. The framework provides class-agnostic hyperedge-importance scores that identify entity configurations and temporal intervals emphasized by the model without treating them as causal explanations. Quantitative evaluation is conducted on EPIC-KITCHENS-100/VISOR and Assembly101 under an annotation-assisted entity-localization protocol. Video-only and entity-based methods provide contextual comparisons, while a matched pairwise graph and a static hypergraph serve as the principal controlled baselines because they use identical entity inputs and comparable relational settings. The proposed method improves HO-F1 over the matched pairwise graph by 6.9 percentage points on EPIC-KITCHENS-100/VISOR and 9.5 points on Assembly101, and exceeds the static hypergraph by 4.4 and 5.8 points, respectively. Qualitative analysis on ARCTIC further shows correspondence between highly ranked hyperedges and contact-rich manipulation intervals. These results demonstrate the value of time-varying higher-order relational modeling for fine-grained manipulation activity recognition.