From Images to 3D Shape Attributes
David F. Fouhey, Abhinav Gupta, Andrew Zisserman

TL;DR
This paper explores how to infer 3D shape properties and embeddings from a single image using CNNs, leveraging synthetic and real sculpture datasets to understand shape cues and generalization capabilities.
Contribution
It introduces a method to predict 3D shape attributes and embeddings from single images, including a large sculpture dataset and analysis of CNN interpretability and generalization.
Findings
CNN identifies key regions for shape inference
Shape embeddings enable sculpture matching across viewpoints
3D shape attributes generalize to various object classes
Abstract
Our goal in this paper is to investigate properties of 3D shape that can be determined from a single image. We define 3D shape attributes -- generic properties of the shape that capture curvature, contact and occupied space. Our first objective is to infer these 3D shape attributes from a single image. A second objective is to infer a 3D shape embedding -- a low dimensional vector representing the 3D shape. We study how the 3D shape attributes and embedding can be obtained from a single image by training a Convolutional Neural Network (CNN) for this task. We start with synthetic images so that the contribution of various cues and nuisance parameters can be controlled. We then turn to real images and introduce a large scale image dataset of sculptures containing 143K images covering 2197 works from 242 artists. For the CNN trained on the sculpture dataset we show the following: (i)…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
Topics3D Shape Modeling and Analysis · Aesthetic Perception and Analysis · Computer Graphics and Visualization Techniques
