OSCAR: Open-Set CAD Retrieval from a Language Prompt and a Single Image

• URL: http://arxiv.org/abs/2601.07333v1
• Date: Mon, 12 Jan 2026 08:59:22 GMT
• Title: OSCAR: Open-Set CAD Retrieval from a Language Prompt and a Single Image
• Authors: Tessa Pulli, Jean-Baptiste Weibel, Peter Hönig, Matthias Hirschmanner, Markus Vincze, Andreas Holzinger,
• Abstract summary: We introduce an Open-Set CAD Retrieval from a Language Prompt and a Single Image (OSCAR)<n>OSCAR retrieves a matching object model from an unlabeled 3D object database.<n>We show that OSCAR achieves an average precision of 90.48% during object retrieval on the YCB-V object dataset.
• Score: 7.903776990851827
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: 6D object pose estimation plays a crucial role in scene understanding for applications such as robotics and augmented reality. To support the needs of ever-changing object sets in such context, modern zero-shot object pose estimators were developed to not require object-specific training but only rely on CAD models. Such models are hard to obtain once deployed, and a continuously changing and growing set of objects makes it harder to reliably identify the instance model of interest. To address this challenge, we introduce an Open-Set CAD Retrieval from a Language Prompt and a Single Image (OSCAR), a novel training-free method that retrieves a matching object model from an unlabeled 3D object database. During onboarding, OSCAR generates multi-view renderings of database models and annotates them with descriptive captions using an image captioning model. At inference, GroundedSAM detects the queried object in the input image, and multi-modal embeddings are computed for both the Region-of-Interest and the database captions. OSCAR employs a two-stage retrieval: text-based filtering using CLIP identifies candidate models, followed by image-based refinement using DINOv2 to select the most visually similar object. In our experiments we demonstrate that OSCAR outperforms all state-of-the-art methods on the cross-domain 3D model retrieval benchmark MI3DOR. Furthermore, we demonstrate OSCAR's direct applicability in automating object model sourcing for 6D object pose estimation. We propose using the most similar object model for pose estimation if the exact instance is not available and show that OSCAR achieves an average precision of 90.48\% during object retrieval on the YCB-V object dataset. Moreover, we demonstrate that the most similar object model can be utilized for pose estimation using Megapose achieving better results than a reconstruction-based approach.

Related papers

• One2Any: One-Reference 6D Pose Estimation for Any Object [98.50085481362808]
  6D object pose estimation remains challenging for many applications due to dependencies on complete 3D models, multi-view images, or training limited to specific object categories.<n>We propose a novel method One2Any that estimates the relative 6-degrees of freedom (DOF) object pose using only a single reference-single query RGB-D image.<n> Experiments on multiple benchmark datasets demonstrate that our model generalizes well to novel objects, achieving state-of-the-art accuracy and even rivaling methods that require multi-view or CAD inputs, at a fraction of compute.
  arXiv  Detail & Related papers  (2025-05-07T03:54:59Z)
• FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects [55.77542145604758]
  FoundationPose is a unified foundation model for 6D object pose estimation and tracking. Our approach can be instantly applied at test-time to a novel object without fine-tuning.
  arXiv  Detail & Related papers  (2023-12-13T18:28:09Z)
• FoundPose: Unseen Object Pose Estimation with Foundation Features [11.32559845631345]
  FoundPose is a model-based method for 6D pose estimation of unseen objects from a single RGB image. The method can quickly onboard new objects using their 3D models without requiring any object- or task-specific training.
  arXiv  Detail & Related papers  (2023-11-30T18:52:29Z)
• LocaliseBot: Multi-view 3D object localisation with differentiable rendering for robot grasping [9.690844449175948]
  We focus on object pose estimation. Our approach relies on three pieces of information: multiple views of the object, the camera's parameters at those viewpoints, and 3D CAD models of objects. We show that the estimated object pose results in 99.65% grasp accuracy with the ground truth grasp candidates.
  arXiv  Detail & Related papers  (2023-11-14T14:27:53Z)
• NOPE: Novel Object Pose Estimation from a Single Image [67.11073133072527]
  We propose an approach that takes a single image of a new object as input and predicts the relative pose of this object in new images without prior knowledge of the object's 3D model. We achieve this by training a model to directly predict discriminative embeddings for viewpoints surrounding the object. This prediction is done using a simple U-Net architecture with attention and conditioned on the desired pose, which yields extremely fast inference.
  arXiv  Detail & Related papers  (2023-03-23T18:55:43Z)
• OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Models [51.68715543630427]
  OnePose relies on detecting repeatable image keypoints and is thus prone to failure on low-textured objects. We propose a keypoint-free pose estimation pipeline to remove the need for repeatable keypoint detection. A 2D-3D matching network directly establishes 2D-3D correspondences between the query image and the reconstructed point-cloud model.
  arXiv  Detail & Related papers  (2023-01-18T17:47:13Z)
• MegaPose: 6D Pose Estimation of Novel Objects via Render & Compare [84.80956484848505]
  MegaPose is a method to estimate the 6D pose of novel objects, that is, objects unseen during training. We present a 6D pose refiner based on a render&compare strategy which can be applied to novel objects. Second, we introduce a novel approach for coarse pose estimation which leverages a network trained to classify whether the pose error between a synthetic rendering and an observed image of the same object can be corrected by the refiner.
  arXiv  Detail & Related papers  (2022-12-13T19:30:03Z)
• OnePose: One-Shot Object Pose Estimation without CAD Models [30.307122037051126]
  OnePose does not rely on CAD models and can handle objects in arbitrary categories without instance- or category-specific network training. OnePose draws the idea from visual localization and only requires a simple RGB video scan of the object to build a sparse SfM model of the object. To mitigate the slow runtime of existing visual localization methods, we propose a new graph attention network that directly matches 2D interest points in the query image with the 3D points in the SfM model.
  arXiv  Detail & Related papers  (2022-05-24T17:59:21Z)
• DONet: Learning Category-Level 6D Object Pose and Size Estimation from Depth Observation [53.55300278592281]
  We propose a method of Category-level 6D Object Pose and Size Estimation (COPSE) from a single depth image. Our framework makes inferences based on the rich geometric information of the object in the depth channel alone. Our framework competes with state-of-the-art approaches that require labeled real-world images.
  arXiv  Detail & Related papers  (2021-06-27T10:41:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.