Geometric Correspondence Fields: Learned Differentiable Rendering for 3D Pose Refinement in the Wild

Alexander Grabner, Yaming Wang, Peizhao Zhang, Peihong Guo, Tong Xiao, Peter Vajda, Peter M. Roth, Vincent Lepetit

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem Konferenzband

Abstract

We present a novel 3D pose refinement approach based ondifferentiable rendering for objects of arbitrary categories in the wild. Incontrast to previous methods, we make two main contributions: First,instead of comparing real-world images and synthetic renderings in theRGB or mask space, we compare them in a feature space optimized for3D pose refinement. Second, we introduce a novel differentiable rendererthat learns to approximate the rasterization backward pass from data in-stead of relying on a hand-crafted algorithm. For this purpose, we predictdeep cross-domain correspondences between RGB images and 3D modelrenderings in the form of what we call geometric correspondence fields.These correspondence fields serve as pixel-level gradients which are ana-lytically propagated backward through the rendering pipeline to performa gradient-based optimization directly on the 3D pose. In this way, weprecisely align 3D models to objects in RGB images which results in sig-nificantly improved 3D pose estimates. We evaluate our approach on thechallenging Pix3D dataset and achieve up to 55% relative improvementcompared to state-of-the-art refinement methods in multiple metrics.
Originalspracheenglisch
TitelEuropean Conference on Computer Vision (ECCV)
Seiten1-14
Seitenumfang14
PublikationsstatusAngenommen/In Druck - 2020
Veranstaltung16th European Conference on Computer Vision: ECCV 2020 - Virtuell, Großbritannien / Vereinigtes Königreich
Dauer: 23 Aug 202028 Aug 2020

Konferenz

Konferenz16th European Conference on Computer Vision
KurztitelECCV 2020
LandGroßbritannien / Vereinigtes Königreich
OrtVirtuell
Zeitraum23/08/2028/08/20

Fingerprint Untersuchen Sie die Forschungsthemen von „Geometric Correspondence Fields: Learned Differentiable Rendering for 3D Pose Refinement in the Wild“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren