Saved in:
Bibliographic Details
Main Authors: Jin, SeongMin, Jeong, Doo Seok
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.11743
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914557625106432
author Jin, SeongMin
Jeong, Doo Seok
author_facet Jin, SeongMin
Jeong, Doo Seok
contents Learning latent representations that capture both semantic and spatial information is central to efficient spatio-semantic reasoning. However, many existing approaches rely on implicit latent structures combined with dense feature maps or task-specific heads, limiting computational efficiency and flexibility. We propose WorldComp2D, a novel lightweight representation learning framework that explicitly structures latent space geometry according to object identity and spatial proximity using multiscale local receptive fields. This framework consists of (i) a proximity-dependent encoder that maps a given observation into a spatio-semantic latent space and (ii) a localizer that infers the coordinates of objects in the input from the resulting spatio-semantic representation. Using facial landmark localization as a proof-of-concept, we show that, compared to SoTA lightweight models, WorldComp2D reduces the numbers of parameters and FLOPs by up to 4.0X and 2.2X, respectively, while maintaining real-time performance on CPU. These results demonstrate that explicitly structured latent spaces provide an efficient and general foundation for spatio-semantic reasoning. This framework is open-sourced at https://github.com/JinSeongmin/WorldComp2D.
format Preprint
id arxiv_https___arxiv_org_abs_2605_11743
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WorldComp2D: Spatio-semantic Representations of Object Identity and Location from Local Views
Jin, SeongMin
Jeong, Doo Seok
Computer Vision and Pattern Recognition
Machine Learning
Learning latent representations that capture both semantic and spatial information is central to efficient spatio-semantic reasoning. However, many existing approaches rely on implicit latent structures combined with dense feature maps or task-specific heads, limiting computational efficiency and flexibility. We propose WorldComp2D, a novel lightweight representation learning framework that explicitly structures latent space geometry according to object identity and spatial proximity using multiscale local receptive fields. This framework consists of (i) a proximity-dependent encoder that maps a given observation into a spatio-semantic latent space and (ii) a localizer that infers the coordinates of objects in the input from the resulting spatio-semantic representation. Using facial landmark localization as a proof-of-concept, we show that, compared to SoTA lightweight models, WorldComp2D reduces the numbers of parameters and FLOPs by up to 4.0X and 2.2X, respectively, while maintaining real-time performance on CPU. These results demonstrate that explicitly structured latent spaces provide an efficient and general foundation for spatio-semantic reasoning. This framework is open-sourced at https://github.com/JinSeongmin/WorldComp2D.
title WorldComp2D: Spatio-semantic Representations of Object Identity and Location from Local Views
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2605.11743