Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 20, 2026

Functional Magnetic Resonance Imaging (fMRI) of the Visual Cortex with Wide-View Retinotopic Stimulation
07:11

Functional Magnetic Resonance Imaging (fMRI) of the Visual Cortex with Wide-View Retinotopic Stimulation

Published on: December 8, 2023

Improving Viewpoint Robustness for Visual Recognition via Adversarial Training.

Shouwei Ruan, Yinpeng Dong, Hang Su

    IEEE Transactions on Pattern Analysis and Machine Intelligence
    |June 18, 2026
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Hierarchical laser-programmed soft actuators for designing bionic robots with freeform morphing shapes.

    Science advances·2026
    Same author

    Robust Image-Based Visual Servoing Formation Control for Quadrotors Without Communication via Reinforcement Learning.

    IEEE transactions on neural networks and learning systems·2026
    Same author

    Input-Constrained Visual Servoing Formation Control for Quadrotors Using Off-Policy Reinforcement Learning.

    IEEE transactions on cybernetics·2026
    Same author

    BPFNN: Bayesian Probabilistic Fuzzy Neural Networks for Uncertainty-Aware Clustering and Probabilistic Fuzzy Reasoning.

    IEEE transactions on cybernetics·2025
    Same author

    Invariant Ellipsoids Method for Homogeneous Leader-Following Consensus Control.

    IEEE transactions on cybernetics·2025
    Same author

    Create Machine Vision Inspired by Eagle Eye.

    Research (Washington, D.C.)·2024
    Same journal

    Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same journal

    RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same journal

    CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same journal

    DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same journal

    Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same journal

    Learning Shape Anchors for Holistic Indoor Scene Understanding.

    IEEE transactions on pattern analysis and machine intelligence·2026
    See all related articles

    We introduce Viewpoint-Invariant Adversarial Training (VIAT) to enhance visual recognition models' robustness against changing viewpoints. This method improves performance across various vision tasks and model architectures.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Artificial Intelligence

    Background:

    • Viewpoint invariance is a critical challenge in 3D visual recognition, as object appearance changes significantly with viewing direction.
    • Existing research primarily focuses on 2D image transformations (translation, rotation), neglecting viewpoint robustness.

    Purpose of the Study:

    • To develop a novel adversarial training method for improving viewpoint robustness in vision models.
    • To address the challenge of viewpoint invariance in 3D visual recognition tasks.

    Main Methods:

    • Proposed Viewpoint-Invariant Adversarial Training (VIAT) as a minimax optimization problem, treating viewpoint changes as adversarial attacks.
    • Introduced GMVFool for generating diverse adversarial viewpoints and ViewRS for certified viewpoint robustness evaluation.

    Related Experiment Videos

    Last Updated: Jun 20, 2026

    Functional Magnetic Resonance Imaging (fMRI) of the Visual Cortex with Wide-View Retinotopic Stimulation
    07:11

    Functional Magnetic Resonance Imaging (fMRI) of the Visual Cortex with Wide-View Retinotopic Stimulation

    Published on: December 8, 2023

  • Developed VIAT-FP (Full Parameter Fine-tuning) and VIAT-PEIT (Parameter-Efficient Instruction-Tuning) for different model scales.

    • Main Results:

    • Significantly improved viewpoint robustness across various vision models, including CNNs, ViTs, and multimodal large language models.
    • Introduced ImageNet-V+, a large-scale dataset for benchmarking viewpoint robustness in image recognition, VQA, and visual entailment.
    • Demonstrated the effectiveness of VIAT in enhancing model performance under diverse viewpoint variations.

    Conclusions:

    • VIAT offers a robust solution for achieving viewpoint invariance in computer vision.
    • The proposed methods and datasets advance the field of viewpoint-robust visual recognition.
    • This work paves the way for more reliable AI systems in real-world 3D environments.