# Joint Optimization for 4D Human-Scene Reconstruction in the Wild

arXiv: https://arxiv.org/abs/2501.02158 · HTML: https://mlsanity.com/paper/2501.02158/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 47 references that are in the map.
- First posted on arXiv 4 Jan 2025.

## Builds on (most cited first, 30 of 38)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [4D Gaussian Splatting for Real-Time Dynamic Scene Rendering](https://mlsanity.com/paper/2310.08528/) (2023)
- [Grounding Image Matching in 3D with MASt3R](https://mlsanity.com/paper/2406.09756/) (2024)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras](https://mlsanity.com/paper/2108.10869/) (2021)
- [ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth](https://mlsanity.com/paper/2302.12288/) (2023)
- [ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation](https://mlsanity.com/paper/2204.12484/) (2022)
- [MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion](https://mlsanity.com/paper/2410.03825/) (2024)
- [Humans in 4D: Reconstructing and Tracking Humans with Transformers](https://mlsanity.com/paper/2305.20091/) (2023)
- [HuMoR: 3D Human Motion Model for Robust Pose Estimation](https://mlsanity.com/paper/2105.04668/) (2021)
- [Resolving 3D Human Pose Ambiguities with 3D Scene Constraints](https://mlsanity.com/paper/1908.06963/) (2019)
- [WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion](https://mlsanity.com/paper/2312.07531/) (2023)
- [Tracking Anything with Decoupled Video Segmentation](https://mlsanity.com/paper/2309.03903/) (2023)
- [Capturing and Inferring Dense Full-Body Human-Scene Contact](https://mlsanity.com/paper/2206.09553/) (2022)
- [MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motion](https://mlsanity.com/paper/2409.19152/) (2024)
- [SMPLer-X: Scaling Up Expressive Human Pose and Shape Estimation](https://mlsanity.com/paper/2309.17448/) (2023)
- [GLAMR: Global Occlusion-Aware Human Mesh Recovery with Dynamic Cameras](https://mlsanity.com/paper/2112.01524/) (2021)
- [Decoupling Human and Camera Motion from Videos in the Wild](https://mlsanity.com/paper/2302.12827/) (2023)
- [Scaling Up Dynamic Human-Scene Interaction Modeling](https://mlsanity.com/paper/2403.08629/) (2024)
- [World-Grounded Human Motion Recovery via Gravity-View Coordinates](https://mlsanity.com/paper/2409.06662/) (2024)
- [TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos](https://mlsanity.com/paper/2403.17346/) (2024)
- [Learning Motion Priors for 4D Human Body Capture in 3D Scenes](https://mlsanity.com/paper/2108.10399/) (2021)
- [EMDB: The Electromagnetic Database of Global 3D Human Pose and Shape in the Wild](https://mlsanity.com/paper/2308.16894/) (2023)
- [TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environments](https://mlsanity.com/paper/2306.02850/) (2023)
- [SLOPER4D: A Scene-Aware Dataset for Global 4D Human Pose Estimation in Urban Environments](https://mlsanity.com/paper/2303.09095/) (2023)
- [4D Human Body Capture from Egocentric Video via 3D Scene Grounding](https://mlsanity.com/paper/2011.13341/) (2020)
- [HSC4D: Human-centered 4D Scene Capture in Large-scale Indoor-outdoor Space Using Wearable IMUs and LiDAR](https://mlsanity.com/paper/2203.09215/) (2022)

## Built on by (most cited first, 29 of 29)

- [Visual Imitation Enables Contextual Humanoid Control](https://mlsanity.com/paper/2505.03729/) (2025)
- [Human3R: Everyone Everywhere All at Once](https://mlsanity.com/paper/2510.06219/) (2025)
- [HAMSt3R: Human-Aware Multi-view Stereo 3D Reconstruction](https://mlsanity.com/paper/2508.16433/) (2025)
- [Reconstructing 4D Spatial Intelligence: A Survey](https://mlsanity.com/paper/2507.21045/) (2025)
- [UniSH: Unifying Scene and Human Reconstruction in a Feed-Forward Pass](https://mlsanity.com/paper/2601.01222/) (2026)
- [Vision Pretraining for Dense Spatial Perception](https://mlsanity.com/paper/2607.05247/) (2026)
- [E3D-Bench: A Benchmark for End-to-End 3D Geometric Foundation Models](https://mlsanity.com/paper/2506.01933/) (2025)
- [DuoMo: Dual Motion Diffusion for World-Space Human Reconstruction](https://mlsanity.com/paper/2603.03265/) (2026)
- [OnlineHMR: Video-based Online World-Grounded Human Mesh Recovery](https://mlsanity.com/paper/2603.17355/) (2026)
- [ACE-Data-0: Human-Centric Ambient Capture as Embodied Data Engine](https://mlsanity.com/paper/2607.28625/) (2026)
- [GRAFT: Geometric Refinement and Fitting Transformer for Human Scene Reconstruction](https://mlsanity.com/paper/2604.19624/) (2026)
- [Inferring Compositional 4D Scenes without Ever Seeing One](https://mlsanity.com/paper/2512.05272/) (2025)
- [Hand3R: Online 4D Hand-Scene Reconstruction in the Wild](https://mlsanity.com/paper/2602.03200/) (2026)
- [EmbodMocap: In-the-Wild 4D Human-Scene Reconstruction for Embodied Agents](https://mlsanity.com/paper/2602.23205/) (2026)
- [End-to-End Spatial-Temporal Transformer for Real-time 4D HOI Reconstruction](https://mlsanity.com/paper/2603.14435/) (2026)
- [UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video](https://mlsanity.com/paper/2604.19923/) (2026)
- [WATCH: World-aware Allied Trajectory and pose reConstruction for Camera and Human](https://mlsanity.com/paper/2509.04600/) (2025)
- [SHARE: Scene-Human Aligned Reconstruction](https://mlsanity.com/paper/2510.15342/) (2025)
- [CRISP: Contact-Guided Real2Sim from Monocular Video with Planar Scene Primitives](https://mlsanity.com/paper/2512.14696/) (2025)
- [HARMONI: Aligning Human and Scene Priors for Multi-View 4D Reconstruction](https://mlsanity.com/paper/2603.12789/) (2026)
- [Pi-HOC: Pairwise 3D Human-Object Contact Estimation](https://mlsanity.com/paper/2604.12923/) (2026)
- [RHINO: Reconstructing Human Interactions with Novel Objects from Monocular Videos](https://mlsanity.com/paper/2605.17014/) (2026)
- [Natural Human Motion Recovery by Aligning High-Order Temporal Dynamics from Monocular Videos](https://mlsanity.com/paper/2605.26879/) (2026)
- [TROPHIES: Temporal Reconstruction of Places, Humans, and Cameras from Multi-view Videos](https://mlsanity.com/paper/2606.02350/) (2026)
- [EgoHTR: Egocentric 4D Demonstrations of Human Terrain Traversal](https://mlsanity.com/paper/2607.13472/) (2026)
- [Crowd4D: Scene-Aware Monocular 4D Crowd Reconstruction](https://mlsanity.com/paper/2607.19517/) (2026)
- [Everybody Tracking Every Body](https://mlsanity.com/paper/2608.29927/) (2026)
- [Feed-Forward Multi-view Multi-person Reconstruction with Contrastive Human-Aware 3D Representation](https://mlsanity.com/paper/2609.00745/) (2026)
- [WildHSR: Metric Feed-Forward 4D People-Scene Reconstruction from a 3D Foundation Model](https://mlsanity.com/paper/2609.29106/) (2026)

## Abstract (arXiv)

Reconstructing human motion and its surrounding environment is crucial for understanding human-scene interaction and predicting human movements in the scene. While much progress has been made in capturing human-scene interaction in constrained environments, those prior methods can hardly reconstruct the natural and diverse human motion and scene context from web videos. In this work, we propose JOSH, a novel optimization-based method for 4D human-scene reconstruction in the wild from monocular videos. JOSH uses techniques in both dense scene reconstruction and human mesh recovery as initialization, and then it leverages the human-scene contact constraints to jointly optimize the scene, the camera poses, and the human motion. Experiment results show JOSH achieves better results on both global human motion estimation and dense scene reconstruction by joint optimization of scene geometry and human motion. We further design a more efficient model, JOSH3R, and directly train it with pseudo-labels from web videos. JOSH3R outperforms other optimization-free methods by only training with labels predicted from JOSH, further demonstrating its accuracy and generalization ability.