What question did this study set out to answer?

This research aims to enhance recognition of deformed 2D shapes using deep learning techniques with projective invariance.

February 26, 2026Neural Computing and ApplicationsOpen Access

Providing projective and affine invariance for recognition by Multi-Angle-Scale Vision Transformer

View Full Paper

Ask AI

Bookmark

Authors

LCLuiz Gustavo da Rocha Charamba NFN. Ferreira SMSilvio de Barros Melo

Discussion

Member takes

Overview

Deep learning demonstrates improved recognition of traffic signs with projective invariance, indicating significant advances in handling geometric deformations.

Key Points

This research aims to enhance recognition of deformed 2D shapes using deep learning techniques with projective invariance.
Introduced MASViT, a deep-learning model for deformed image recognition.
Employed 1D convolutional filters for shape representation in the polar domain.
Implemented regularization techniques to improve generalizability of the model.
Validated the approach with curated datasets derived from the GTSRB dataset.
The approach outperformed state-of-the-art methods in recognizing affinely and projectively deformed images.
Demonstrated enhanced performance particularly for severe geometric deformations.

Cite This Study

Charamba et al. (2026) studied this question.

synapsesocial.com/papers/699f95951bc9fecf3dab395a https://doi.org/10.1007/s00521-025-11821-2

View Full Paper

Ask AI

Bookmark

February 26, 2026Neural Computing and ApplicationsOpen Access

Providing projective and affine invariance for recognition by Multi-Angle-Scale Vision Transformer

View Full Paper

Ask AI

Bookmark

Authors

LCLuiz Gustavo da Rocha Charamba NFN. Ferreira SMSilvio de Barros Melo

Discussion

Member takes

Overview

Deep learning demonstrates improved recognition of traffic signs with projective invariance, indicating significant advances in handling geometric deformations.

Key Points

This research aims to enhance recognition of deformed 2D shapes using deep learning techniques with projective invariance.
Introduced MASViT, a deep-learning model for deformed image recognition.
Employed 1D convolutional filters for shape representation in the polar domain.
Implemented regularization techniques to improve generalizability of the model.
Validated the approach with curated datasets derived from the GTSRB dataset.
The approach outperformed state-of-the-art methods in recognizing affinely and projectively deformed images.
Demonstrated enhanced performance particularly for severe geometric deformations.

Cite This Study

Charamba et al. (2026) studied this question.

synapsesocial.com/papers/699f95951bc9fecf3dab395a https://doi.org/10.1007/s00521-025-11821-2

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark