From Local Cues to Global Percepts: Emergent Gestalt Organization in Self-Supervised Vision Models

• URL: http://arxiv.org/abs/2506.00718v1
• Date: Sat, 31 May 2025 21:35:54 GMT
• Title: From Local Cues to Global Percepts: Emergent Gestalt Organization in Self-Supervised Vision Models
• Authors: Tianqin Li, Ziqi Wen, Leiran Song, Jun Liu, Zhi Jing, Tai Sing Lee,
• Abstract summary: We investigate whether modern vision models show similar behaviors, and under what training conditions these emerge.<n>We find that Vision Transformers (ViTs) trained with Masked Autoencoding (MAE) exhibit activation patterns consistent with Gestalt laws.<n>We introduce the Distorted Spatial Relationship Testbench (DiSRT), which evaluates sensitivity to global spatial perturbations while preserving local textures.
• Score: 7.7536110932446265
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Human vision organizes local cues into coherent global forms using Gestalt principles like closure, proximity, and figure-ground assignment -- functions reliant on global spatial structure. We investigate whether modern vision models show similar behaviors, and under what training conditions these emerge. We find that Vision Transformers (ViTs) trained with Masked Autoencoding (MAE) exhibit activation patterns consistent with Gestalt laws, including illusory contour completion, convexity preference, and dynamic figure-ground segregation. To probe the computational basis, we hypothesize that modeling global dependencies is necessary for Gestalt-like organization. We introduce the Distorted Spatial Relationship Testbench (DiSRT), which evaluates sensitivity to global spatial perturbations while preserving local textures. Using DiSRT, we show that self-supervised models (e.g., MAE, CLIP) outperform supervised baselines and sometimes even exceed human performance. ConvNeXt models trained with MAE also exhibit Gestalt-compatible representations, suggesting such sensitivity can arise without attention architectures. However, classification finetuning degrades this ability. Inspired by biological vision, we show that a Top-K activation sparsity mechanism can restore global sensitivity. Our findings identify training conditions that promote or suppress Gestalt-like perception and establish DiSRT as a diagnostic for global structure sensitivity across models.

Related papers

• TARDIS STRIDE: A Spatio-Temporal Road Image Dataset and World Model for Autonomy [44.85881816317044]
  We show how to permuting 360-degree panoramic imagery into rich interconnected observation, state and action nodes.<n>We benchmark this dataset via TARDIS, a transformer-based generative world model.<n>We demonstrate robust performance across a range of agentic tasks such as controllable image synthesis, instruction following, autonomous self-control, and state-of-the-art georeferencing.
  arXiv  Detail & Related papers  (2025-06-12T21:08:11Z)
• TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observation [65.74990259650984]
  We introduce TerraFM, a scalable self-supervised learning model that leverages globally distributed Sentinel-1 and Sentinel-2 imagery.<n>Our training strategy integrates local-global contrastive learning and introduces a dual-centering mechanism.<n>TerraFM achieves strong generalization on both classification and segmentation tasks, outperforming prior models on GEO-Bench and Copernicus-Bench.
  arXiv  Detail & Related papers  (2025-06-06T17:59:50Z)
• SPARTAN: A Sparse Transformer Learning Local Causation [63.29645501232935]
  Causal structures play a central role in world models that flexibly adapt to changes in the environment. We present the SPARse TrANsformer World model (SPARTAN), a Transformer-based world model that learns local causal structures between entities in a scene. By applying sparsity regularisation on the attention pattern between object-factored tokens, SPARTAN identifies sparse local causal models that accurately predict future object states.
  arXiv  Detail & Related papers  (2024-11-11T11:42:48Z)
• A Theoretical Analysis of Self-Supervised Learning for Vision Transformers [66.08606211686339]
  Masked autoencoders (MAE) and contrastive learning (CL) capture different types of representations.<n>We study the training dynamics of one-layer softmax-based vision transformers (ViTs) on both MAE and CL objectives.
  arXiv  Detail & Related papers  (2024-03-04T17:24:03Z)
• Self-Supervised Learning for Place Representation Generalization across Appearance Changes [11.030196234282675]
  We investigate learning features that are robust to appearance modifications while sensitive to geometric transformations in a self-supervised manner. Our results reveal that jointly learning appearance-robust and geometry-sensitive image descriptors leads to competitive visual place recognition results.
  arXiv  Detail & Related papers  (2023-03-04T10:14:47Z)
• Style-Hallucinated Dual Consistency Learning: A Unified Framework for Visual Domain Generalization [113.03189252044773]
  We propose a unified framework, Style-HAllucinated Dual consistEncy learning (SHADE), to handle domain shift in various visual tasks. Our versatile SHADE can significantly enhance the generalization in various visual recognition tasks, including image classification, semantic segmentation and object detection.
  arXiv  Detail & Related papers  (2022-12-18T11:42:51Z)
• Bridging the Gap to Real-World Object-Centric Learning [66.55867830853803]
  We show that reconstructing features from models trained in a self-supervised manner is a sufficient training signal for object-centric representations to arise in a fully unsupervised way. Our approach, DINOSAUR, significantly out-performs existing object-centric learning models on simulated data.
  arXiv  Detail & Related papers  (2022-09-29T15:24:47Z)
• Understanding Dynamics of Nonlinear Representation Learning and Its Application [12.697842097171119]
  We study the dynamics of implicit nonlinear representation learning. We show that the data-architecture alignment condition is sufficient for the global convergence. We derive a new training framework, which satisfies the data-architecture alignment condition without assuming it.
  arXiv  Detail & Related papers  (2021-06-28T16:31:30Z)
• S2RMs: Spatially Structured Recurrent Modules [105.0377129434636]
  We take a step towards exploiting dynamic structure that are capable of simultaneously exploiting both modular andtemporal structures. We find our models to be robust to the number of available views and better capable of generalization to novel tasks without additional training.
  arXiv  Detail & Related papers  (2020-07-13T17:44:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.