A Theory-Inspired Framework for Few-Shot Cross-Modal Sketch Person Re-Identification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gong, Yunpeng, Hou, Yongjie, Shi, Jiangming, Diep, Kim Long, Jiang, Min
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917100316000256
author Gong, Yunpeng
Hou, Yongjie
Shi, Jiangming
Diep, Kim Long
Jiang, Min
author_facet Gong, Yunpeng
Hou, Yongjie
Shi, Jiangming
Diep, Kim Long
Jiang, Min
contents Sketch based person re-identification aims to match hand-drawn sketches with RGB surveillance images, but remains challenging due to significant modality gaps and limited annotated data. To address this, we introduce KTCAA, a theoretically grounded framework for few-shot cross-modal generalization. Motivated by generalization theory, we identify two key factors influencing target domain risk: (1) domain discrepancy, which quantifies the alignment difficulty between source and target distributions; and (2) perturbation invariance, which evaluates the model's robustness to modality shifts. Based on these insights, we propose two components: (1) Alignment Augmentation (AA), which applies localized sketch-style transformations to simulate target distributions and facilitate progressive alignment; and (2) Knowledge Transfer Catalyst (KTC), which enhances invariance by introducing worst-case perturbations and enforcing consistency. These modules are jointly optimized under a meta-learning paradigm that transfers alignment knowledge from data-rich RGB domains to sketch-based scenarios. Experiments on multiple benchmarks demonstrate that KTCAA achieves state-of-the-art performance, particularly in data-scarce conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18677
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Theory-Inspired Framework for Few-Shot Cross-Modal Sketch Person Re-Identification
Gong, Yunpeng
Hou, Yongjie
Shi, Jiangming
Diep, Kim Long
Jiang, Min
Computer Vision and Pattern Recognition
Sketch based person re-identification aims to match hand-drawn sketches with RGB surveillance images, but remains challenging due to significant modality gaps and limited annotated data. To address this, we introduce KTCAA, a theoretically grounded framework for few-shot cross-modal generalization. Motivated by generalization theory, we identify two key factors influencing target domain risk: (1) domain discrepancy, which quantifies the alignment difficulty between source and target distributions; and (2) perturbation invariance, which evaluates the model's robustness to modality shifts. Based on these insights, we propose two components: (1) Alignment Augmentation (AA), which applies localized sketch-style transformations to simulate target distributions and facilitate progressive alignment; and (2) Knowledge Transfer Catalyst (KTC), which enhances invariance by introducing worst-case perturbations and enforcing consistency. These modules are jointly optimized under a meta-learning paradigm that transfers alignment knowledge from data-rich RGB domains to sketch-based scenarios. Experiments on multiple benchmarks demonstrate that KTCAA achieves state-of-the-art performance, particularly in data-scarce conditions.
title A Theory-Inspired Framework for Few-Shot Cross-Modal Sketch Person Re-Identification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.18677