RYOPO is an end‑to‑end query‑based RGB‑D set predictor that jointly detects, segments, and estimates 9‑DoF poses of unseen instances within known categories without relying on external instance segmentation or CAD priors. It uses shared image and scene encoding, a query‑conditioned geometry pathway, and object‑centric refinement with pose‑conditioned cross‑attention to achieve accurate pose estimation. On benchmark datasets such as NOCS, REAL275, and HouseCat6D, RYOPO outperforms published methods and runs in real time at 31.8 FPS on an RTX A6000.
By Hakjin Lee, Junghoon Seo, Jaehoon Sim
MessyKitchens introduces a new dataset of cluttered real-world kitchen scenes with detailed 3D object shapes, poses, and accurate contact information. The authors extend the SAM 3D single-object reconstruction method with a Multi-Object Decoder (MOD) to jointly reconstruct entire scenes, achieving better registration accuracy and reduced inter-object penetration compared to prior work. The dataset, benchmark, code, and pretrained models will be publicly released on the project website.
By Junaid Ahmed Ansari, Ran Ding, Fabio Pizzati, Ivan Laptev
arXiv:2609.23796v2 Announce Type: replace
Abstract: Single-image 3D object generation can now produce high-fidelity assets, yet accurately placing them into a coherent scene layout remains an open ch...
By Yang-Tian Sun, Tianjia Liu, Zehuan Huang, Yi-Hua Huang, Xiaoyang Lyu, Ziyi Yang, Zi-Xin Zou, Yuan-Chen Guo, Yan-Pei Cao, Xiaojuan Qi
SceneReGen is a new framework for reconstructing 3D scenes from a single image by generating and assembling complete object meshes within a shared observation‑aligned scene frame. It uses selective pose factorization to encode each object’s observed orientation directly into the generated mesh, while estimating translation and scale from instance‑level and global scene cues. Evaluated on the 3D‑FUTURE dataset, SceneReGen outperforms existing methods on scene‑level metrics and shows strong performance on object‑level metrics, demonstrating its effectiveness in autonomous‑driving and embodied‑AI scenarios.
By Zefan Tian, Yuteng Ye, Yiheng Zhang, Yuhang Yang, Xueqiang Lv, Shizhou Zhang, Le Liu, Di Xu
arXiv:2609.23796v1 Announce Type: new
Abstract: Single-image 3D object generation can now produce high-fidelity assets, yet accurately placing them into a coherent scene layout remains an open challe...
By Yang-Tian Sun, Tianjia Liu, Zehuan Huang, Yi-Hua Huang, Xiaoyang Lyu, Ziyi Yang, Zi-Xin Zou, Yuan-Chen Guo, Yan-Pei Cao, Xiaojuan Qi
The paper presents a generative framework that estimates category-level 6D pose and 3D size of objects from a single RGB image, using score-based diffusion models to produce a multi-hypothesis pose distribution. It replaces costly likelihood pruning with a Mean Shift approach to isolate the mode as the final pose estimate, achieving state-of-the-art results on the REAL275 benchmark. The method also decouples detection from pose estimation, enabling robust zero-shot generalisation on the Wild6D dataset and extending naturally to video sequences by propagating the pose distribution over time.
By Adam Bethell, Ravi Garg, Ian Reid
F2G-Pose is a real‑time, single‑pass framework for category‑level RGB‑D object pose estimation that predicts 3D rotation, translation, metric size, and a completed camera‑frame shape from a single RGB‑D observation. It lifts dense visual foundation features to partial point clouds, encodes local geometry with DGCNN tokenization, and uses a Mixture‑of‑Experts Transformer to integrate object‑level structure. Trained only on pose‑annotated synthetic SOPE data and leveraging a frozen RADIO backbone, F2G‑Pose achieves state‑of‑the‑art accuracy on SOPE and ROPE while running at 28 FPS for pose‑only inference and demonstrates strong geometry completion on HouseCat6D and real‑object transfer on HANDAL.
By Jinyu Zhang, Haitao Lin, Jiashu Hou, Xiangyang Xue, Yanwei Fu
arXiv:2607.09985v2 Announce Type: replace
Abstract: Object pose estimation is a fundamental problem in 3D vision. Although recent state-of-the-art approaches achieve strong performance, generalizatio...
By Yang You, Yi Du, Cole Harrison, Leonidas Guibas
LucidLyra is a framework that reconstructs 3D scenes from video latents in a feed‑forward manner while simultaneously providing open‑vocabulary semantics and instance cues. By distilling knowledge from a camera‑controlled video diffusion model and frozen 2D vision foundation models, it jointly infers 3D Gaussians that encode geometry, appearance, and semantic information without requiring per‑scene optimization or manual 3D annotations. The resulting scenes support novel‑view rendering, open‑vocabulary segmentation, and object grouping, and can be extended to monocular video for dynamic 3D reconstruction.
Dyna3 is a training‑free framework that extends the depth foundation model DA3 to perform 4D dynamic scene reconstruction without fine‑tuning. By leveraging DA3’s cross‑view features and a best‑match search, it distinguishes static surfaces from moving objects, and uses vision‑language models to generate semantic prompts for SAM 3 to achieve precise instance‑level segmentation. Experiments on four datasets show Dyna3 outperforms correspondence‑trained methods, improving dynamic object segmentation by +5.5 pp, speeding pose estimation 13×, and reducing memory usage 4–8×.
By Xinhao Xiang, Weiyang Li, Zhijie Zheng, Abhijeet Rastogi, Jiawei Zhang
GenCOPE introduces a synthetic-to-real (Syn2Real) approach for category-level object pose estimation (COPE) that eliminates the need for labor-intensive real-world data collection. By learning domain-invariant representations through 2D and 3D semantic consistency constraints and employing an end-to-end pose regression framework with 2D-3D cross consistency, the model achieves robust generalization across synthetic and real domains. The architecture relies solely on global features, resulting in a lightweight and efficient design validated on REAL275, Wild6D, and real-world robotic manipulation scenes.
By Jian Liu, Wei Sun, Zhenqi Dai, Hui Yang, Jian Xiao, Nicu Sebe, Na Zhao
The paper introduces a biologically inspired framework that learns object‑centric visual representations from raw videos without human annotations or camera calibration. By using motion boundaries detected via optical flow and clustering to create pseudo‑instance masks, the method supervises a single‑image encoder with pixel‑level pairwise metric learning. Training on 195 million pseudo‑labeled frames and expanding to 421 million frames through Motion‑Verified Self‑Training, the approach yields Swin‑based encoders that outperform or match supervised and self‑supervised baselines on tasks such as monocular depth estimation, 3D object detection, 3D occupancy prediction, and end‑to‑end planning.
By Boshi Li, Xiaohui Wang, Xiaoyang Wu, Zhichao Li, Ya Yang, Naiyan Wang