AI Paper Atlas

Research map / 3D vision / Face and shape

LoST: Level of Semantics Tokenization for 3D Shapes

Niladri Shekhar Dutt, Zifan Shi, Paul Guerrero, Chun-Hao Paul Huang, Duygu Ceylan, Niloy J. Mitra, Xuelin Chen

CVPR 2026 · 3D vision › Face and shape · approach: 3D shape generative modeling · 2 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
add RIDA semantic alignment loss (ablation)LoST (1 token) · Tokenizer reconstruction (1k shapes)DINO similarity: 0.720 -> 0.731Improved
add RIDA semantic alignment loss (ablation)LoST (512 tokens) · Tokenizer reconstruction (1k shapes)DINO similarity: 0.904 -> 0.921Improved
add RIDA semantic alignment loss (ablation)LoST (16 tokens) · Tokenizer reconstruction (1k shapes)DINO similarity: 0.821 -> 0.814Hurt
add RIDA semantic alignment loss (ablation)LoST (1 token) · Tokenizer reconstruction (1k shapes)DINOv2 similarity: 0.528 -> 0.556Improved
add RIDA semantic alignment loss (ablation)LoST (64 tokens) · Tokenizer reconstruction (1k shapes)DINOv2 similarity: 0.763 -> 0.805Improved
add RIDA semantic alignment loss (ablation)LoST · Tokenizer reconstruction (1k shapes)Chamfer DistanceNo clear effect
add RIDA semantic alignment loss (ablation)LoST diffusion decoder · training convergencetraining speed: ~40% fasterImproved
use RIDA features instead of raw triplane features for retrievalsemantic shape retrieval · Retrieval OOD setmAP@3: 17.47% -> 28.28%Improved
use RIDA features instead of raw triplane features for retrievalsemantic shape retrieval · Retrieval in-distribution setmAP@3: 16.42% -> 44.28%Improved
use RIDA instead of direct MSE regression of DINO features (ablation)semantic extractor · Retrieval in-distribution setmAP@3: 22.90% -> 44.28%Improved
use RIDA instead of direct MSE regression of DINO features (ablation)semantic extractor · Retrieval OOD setRecall@3: 20.63% -> 32.03%Improved
use LoST tokenizer on TRELLIS latents instead of Direct3D triplane latentsLoST (1 token) · Objaverse (128 assets)DINO similarity: 0.690 -> 0.631Hurt

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