Detailed 3D Representations for Object Recognition and Modeling

By M. Zia, Michael Stark, Bernt Schiele, Konrad Schindler. 2013

191 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, 3D Vision & Pose

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (2017, background): “Given that images provide detailed texture information, many algorithms infered the 3D bounding boxes from 2D images [4, 3, 44, 45, 46, 38].”
Multi-view 3D Object Detection Network for Autonomous Driving (2016, methodology): “[27, 28] introduces a detailed geometry representation of objects using 3D wireframe models.”
Lifting Object Detection Datasets into 3D (2015, positive): “To account for differences between the 3D exemplars and the object depicted, Su and Guibas [26] choose a few exemplars that appear to be similar to the depicted shape and combine them to produce a single depth map.”
Category-specific object reconstruction from a single image (2014, methodology): “Here we pursue more expressive basis shape models [1, 7, 42] which establish a balance between the two extremes as they can deform but only along class-specific modes of variation.”
Single Image 3D Interpreter Network (2016, background): “Typically, researchers recover a 3D skeleton from a single image by minimizing its projection error on the 2D image plane [26,27,28].”
Learning 3D Object Categories by Looking Around Them (2017, methodology): “Morphable models have sometimes been used [45, 16], particularly for modeling faces [3, 20].”
SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (2017, methodology): “Estimates of wireframe for 3D objects are obtained by a 3D geometric object class model in [39].”
Are Cars Just 3D Boxes? Jointly Estimating the 3D Shape of Multiple Objects (2014, methodology): “In this paper, we approach the scene understanding problem from a different angle: instead of building a scene model around an off-the-shelf 2D bounding box detector, we start directly from a fine-grained 3D object class model [39], and extend it to jointly represent scenes containing multiple objec”
Learning 3D Part Detection from Sparsely Labeled Data (2014, methodology): “Human-selected vertices in CAD models have been used to train deformable models for pose estimation in images [30].”
GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving (2019, background): “Early works [26, 6, 29, 9, 5] use low-level feature or statistics analysis to tackle 3D recognition or recover tasks.”
3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare (2018, methodology): “When it comes to shape, most methods either estimate only 3D bounding boxes [36, 6, 12], or coarse wire-frame skeletons [29, 54, 63, 62], or represent shape via an exemplar mesh chosen from a small set of meshes [4, 56, 35, 2].”
Sparse Representation for 3D Shape Estimation: A Convex Relaxation Approach (2015, methodology): “[5], [6]) adopted a 3D deformable shape model originated from the “active shape model” [7] to represent shapes, where each shape is defined by a set of ordered landmarks and the shape to be estimated is assumed as a linear combination of predefined basis shapes.”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, methodology): “Zia et al. (2013) exploit the fact that high-quality 3D CAD models are available for many important classes.”
Joint 3d Estimation of Vehicles and Scene Flow (2015, methodology): “(Zia et al., 2013a, Zia et al., 2013b, Zia et al., 2015) revisited the idea of the ASM and applied it to a set of manually annotated CAD models to derive detailed 3D geometric object class representations.”
Joint 3D Object and Layout Inference from a Single RGB-D Image (2015, methodology): “While 3D CAD models have been primarily used for object detection [24, 35, 39,48], holistic 3D scene understanding approaches typically rely on simpler”
Seeing 3D Chairs: Exemplar Part-Based 2D-3D Alignment Using a Large Dataset of CAD Models (2014, background): “In category-level recognition, recent work has explored recognition, alignment and fine pose estimation for cars and bicycles in outdoor scenes using low-dimensional parametric deformable 3D models combined with a relatively small number of learnt part detectors [20, 30, 41].”
Enriching object detection with 2D-3D registration and continuous viewpoint estimation (2015, methodology): “One such method from [29] aligns a 3D deformable part-based wireframe model with input images to accurately predict object poses.”
Deep Exemplar 2D-3D Detection by Adapting from Real to Rendered Views (2015, methodology): “Furthermore, simplified or parametric geometric models have been used for category recognition/detection [15, 20, 24, 42, 58, 60].”