AI Paper Atlas

Research map / 3D vision / Face and shape

Interpretable 3D Neural Object Volumes for Robust Conceptual Reasoning

Nhi Pham, Artur Jesslen, Bernt Schiele, Adam Kortylewski, Jonas Fischer

ICLR 2026 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace CAVE with ground-truth pose by weak Orient-Anything pose supervision (ablation)CAVE · OOD-CVaccuracy (%): 84.0 -> 80.4Hurt
replace CAVE with ground-truth pose by weak Orient-Anything pose supervision (ablation)CAVE · Occluded P3D+accuracy (%): 81.5 -> 77.4Hurt
replace CAVE with ground-truth pose by weak Orient-Anything pose supervision (ablation)CAVE · ImageNet3Daccuracy (%): 88.5 -> 84.7Hurt
replace dense NOVUM Gaussians (about 1130 per class) with 20 K-means concepts per class (ablation)NOVUM · OOD-CV (weak pose)accuracy (%): 79.7 -> 80.4Improved
replace dense NOVUM Gaussians with 20 concepts per class, ground-truth pose (ablation)NOVUM · Pascal3D+accuracy (%): 99.5 -> 99.4No clear effect
replace dense NOVUM Gaussians with 20 concepts per class, weak pose (ablation)NOVUM · ImageNet3Daccuracy (%): 85.7 -> 84.7Hurt
use CAVE concept-based NOVs instead of post-hoc concepts on NOVUMNOVUM · OOD-CV3D consistency: 0.15 -> 0.24Improved
use CAVE instead of strongest prototype network TesNetTesNet · ImageNet3Daccuracy (%): 77.9 -> 84.7Improved
use CAVE instead of LF-CBMLF-CBM · OOD-CVaccuracy (%): 73.5 -> 80.4Improved
use CAVE instead of LF-CBMLF-CBM · Pascal-Partobject coverage: 0.56 -> 0.80Improved
use NOV-aware LRP instead of vanilla LRP or GradCAM (ablation)CAVE concept attribution · qualitative OOD (snow, occlusion)attribution coherenceImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy