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.
arXiv:2605.13018v2 Announce Type: replace Abstract: Object-centric scene understanding is a fundamental challenge in computer vision. Existing approaches often rely on multi-stage pipelines that firs...
SDPose leverages pre‑trained diffusion models, specifically the Stable Diffusion U‑Net, to extract robust latent features for human pose estimation. By identifying the most effective upsampling levels, fusing these features with a lightweight aggregation module, and jointly training with keypoint heatmap supervision and a latent reconstruction loss, SDPose achieves strong cross‑domain generalization. The authors introduce the COCO‑OOD benchmark, comprising style‑transferred and corrupted subsets, to evaluate robustness, and demonstrate that SDPose matches Sapiens on COCO while outperforming competitors in zero‑shot cross‑domain scenarios.
The paper introduces a new approach to 3D object pose estimation that leverages diffusion model features, such as those from Stable Diffusion, to improve performance on unseen objects. Three aggregation architectures—vanilla, nonlinear, and context‑aware—are proposed, with a confidence‑adaptive aggregation network that selectively uses discriminative features to balance speed and accuracy. The method achieves state‑of‑the‑art accuracy on unseen object benchmarks (97.7% vs. 93.5% on Unseen LM) and real‑time performance (68 FPS) while reducing computational cost.
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.
arXiv:2601. 06550v3 Announce Type: replace-cross Abstract: Semantic Multi-Object Tracking (SMOT) is evolving from purely geometric localization toward comprehensive video understanding.
The paper introduces PointVGGT, a zero‑shot framework for multiview RGB‑D point cloud registration that replaces the traditional pairwise‑then‑global pipeline. It employs a foundation‑then‑refinement paradigm, using visual geometry foundation models to directly recover metrically consistent global poses and then refining them with voxelized spatial hashing and IRLS‑based bundle adjustment. Experiments on indoor, object‑centric, and outdoor datasets demonstrate superior registration accuracy and computational efficiency without any training.
CARI4D is a category‑agnostic method that reconstructs spatially and temporally consistent 4D human‑object interactions from monocular RGB videos at metric scale. It introduces a pose hypothesis selection algorithm that integrates predictions from foundation models, refines them via a learned render‑and‑compare paradigm, and reasons about contacts to satisfy physical constraints. Experiments demonstrate a 38% improvement on in‑distribution data and 36% on unseen data compared to prior work, and the model generalizes to unseen categories, enabling zero‑shot application to internet videos.
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...
The paper introduces PointVGGT, a zero‑shot framework for multiview RGB‑D point cloud registration that avoids pairwise estimation by using a foundation‑then‑refinement paradigm. In the foundation stage, it recovers metrically consistent global poses by grounding scale‑ambiguous predictions from visual geometry foundation models against metric depth. The refinement stage employs voxelized spatial hashing and IRLS‑based bundle adjustment to efficiently refine poses, achieving high accuracy and computational efficiency across indoor, object‑centric, and outdoor datasets.
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...
arXiv:2603.12064v3 Announce Type: replace Abstract: We address the challenging problem of dense dynamic scene reconstruction and camera pose estimation from multiple freely moving cameras -- a settin...
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.