Understanding Pixel-level 2D Image Semantics with 3D Keypoint Knowledge Engine

TL;DR

This paper introduces a novel approach for pixel-level 2D image semantics by leveraging 3D keypoint knowledge, enabling explicit reasoning about object visibility and occlusion, and achieving superior results on benchmarks.

Contribution

The paper presents a method that predicts 3D semantic labels and projects them onto 2D images, utilizing a large-scale 3D keypoint dataset to improve semantic understanding.

Findings

Achieves superior results on semantic benchmarks.

Effectively reasons about object occlusion and visibility.

Leverages 3D knowledge for improved 2D semantic understanding.

Abstract

Pixel-level 2D object semantic understanding is an important topic in computer vision and could help machine deeply understand objects (e.g. functionality and affordance) in our daily life. However, most previous methods directly train on correspondences in 2D images, which is end-to-end but loses plenty of information in 3D spaces. In this paper, we propose a new method on predicting image corresponding semantics in 3D domain and then projecting them back onto 2D images to achieve pixel-level understanding. In order to obtain reliable 3D semantic labels that are absent in current image datasets, we build a large scale keypoint knowledge engine called KeypointNet, which contains 103,450 keypoints and 8,234 3D models from 16 object categories. Our method leverages the advantages in 3D vision and can explicitly reason about objects self-occlusion and visibility. We show that our method gives comparative and even superior results on standard semantic benchmarks.