# From Where to How: Continuous 4D Interaction Forecasting from Egocentric Video

arXiv: https://arxiv.org/abs/2609.08636 · HTML: https://mlsanity.com/paper/2609.08636/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 49 of its 70 references that are in the map.
- First posted on arXiv 8 Sep 2026.

## Builds on (most cited first, 30 of 49)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model](https://mlsanity.com/paper/2208.15001/) (2022)
- [Detecting Twenty-thousand Classes using Image-level Supervision](https://mlsanity.com/paper/2201.02605/) (2022)
- [Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives](https://mlsanity.com/paper/2311.18259/) (2023)
- [LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels](https://mlsanity.com/paper/2603.19312/) (2026)
- [The EPIC-KITCHENS Dataset: Collection, Challenges and Baselines](https://mlsanity.com/paper/2005.00343/) (2020)
- [WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion](https://mlsanity.com/paper/2312.07531/) (2023)
- [V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning](https://mlsanity.com/paper/2603.14482/) (2026)
- [Joint Hand Motion and Interaction Hotspots Prediction from Egocentric Videos](https://mlsanity.com/paper/2204.01696/) (2022)
- [Human Motion Generation: A Survey](https://mlsanity.com/paper/2307.10894/) (2023)
- [MotionLLM: Understanding Human Behaviors from Human Motions and Videos](https://mlsanity.com/paper/2405.20340/) (2024)
- [BeLFusion: Latent Diffusion for Behavior-Driven Human Motion Prediction](https://mlsanity.com/paper/2211.14304/) (2022)
- [GIMO: Gaze-Informed Human Motion Prediction in Context](https://mlsanity.com/paper/2204.09443/) (2022)
- [Spatiotemporal Co-attention Recurrent Neural Networks for Human-Skeleton Motion Prediction](https://mlsanity.com/paper/1909.13245/) (2019)
- [Diverse Human Motion Prediction Guided by Multi-Level Spatial-Temporal Anchors](https://mlsanity.com/paper/2302.04860/) (2023)
- [Uncertainty-aware State Space Transformer for Egocentric 3D Hand Trajectory Forecasting](https://mlsanity.com/paper/2307.08243/) (2023)
- [Investigating Pose Representations and Motion Contexts Modeling for 3D Motion Prediction](https://mlsanity.com/paper/2112.15012/) (2021)
- [Egocentric Prediction of Action Target in 3D](https://mlsanity.com/paper/2203.13116/) (2022)
- [Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning](https://mlsanity.com/paper/2404.05218/) (2024)
- [Human Joint Kinematics Diffusion-Refinement for Stochastic Motion Prediction](https://mlsanity.com/paper/2210.05976/) (2022)
- [Satori: Towards Proactive AR Assistant with Belief-Desire-Intention User Modeling](https://mlsanity.com/paper/2410.16668/) (2024)
- [MoGaze: A Dataset of Full-Body Motions that Includes Workspace Geometry and Eye-Gaze](https://mlsanity.com/paper/2011.11552/) (2020)
- [CoMusion: Towards Consistent Stochastic Human Motion Prediction via Motion Diffusion](https://mlsanity.com/paper/2305.12554/) (2023)
- [Diff-IP2D: Diffusion-Based Hand-Object Interaction Prediction on Egocentric Videos](https://mlsanity.com/paper/2405.04370/) (2024)
- [MADiff: Motion-Aware Mamba Diffusion Models for Hand Trajectory Prediction on Egocentric Videos](https://mlsanity.com/paper/2409.02638/) (2024)
- [The Invisible EgoHand: 3D Hand Forecasting through EgoBody Pose Estimation](https://mlsanity.com/paper/2504.08654/) (2025)

## Abstract (arXiv)

Egocentric 4D interaction forecasting aims to anticipate both where future interactions will occur in 3D and how the human body will move to realize them, providing an important capability for assistive robotics and human-computer interaction. Existing methods struggle to translate semantic understanding into precise continuous 3D localization and to balance motion diversity with structural consistency in pose forecasting. More fundamentally, these tasks are often modeled separately, leaving the continuous geometric and temporal correspondence between interaction locations and body motion insufficiently captured. To address these challenges, we introduce Coherent4D, a large-scale egocentric dataset for continuous 4D interaction forecasting, comprising approximately 233K samples across three domains. Each sample pairs a sequence of future 3D interaction locations with corresponding full-body poses, aligned in time and expressed in a shared coordinate system. We also provide evaluation metrics in continuous space. Building on this formulation, we propose HIGFlow, a Hand Interaction Guided Residual Flow framework that models forecasting as a cascaded where-to-how process. HIGFlow first forecasts continuous future interaction locations by combining semantic grounding with short-horizon visual dynamics, and then uses the predicted location sequence to condition a deterministic motion anchor and residual Flow Matching for diverse yet structurally consistent full-body motion forecasting. Extensive experiments across all three domains demonstrate consistent improvements over representative baselines on both location and pose forecasting, while ablations validate the contributions of the proposed components. The project page is available at https://corrineqiu.github.io/from-where-to-how/.