paper-with-me

Papers

Learning Loss for Test-Time Augmentation

2020-10-22 · NeurIPS 2020 12 · Ildoo Kim, Younghoon Kim, Sungwoong Kim

Data augmentation has been actively studied for robust neural networks. Most of the recent data augmentation methods focus on augmenting datasets during the training phase. At the testing phase, simple transformations are still widely used for test-time augmentation. This paper proposes a novel instance-level test-time augmentation that efficiently selects suitable transformations for a test input. Our proposed method involves an auxiliary module to predict the loss of each possible transformation given the input. Then, the transformations having lower predicted losses are applied to the input. The network obtains the results by averaging the prediction results of augmented inputs. Experimental results on several image classification benchmarks show that the proposed instance-aware test-time augmentation improves the model's robustness against various corruptions.

📄 PDF Abstract BibTeX arXiv:2010.11422

Code (0)

등록된 구현이 없습니다.

Tasks

Data Augmentationimage-classificationImage Classification

Similar Papers 제목 키워드 기반

Cyclic Test Time Augmentation with Entropy Weight Method

2021-09-29 · Sewhan Chun, Jae Young Lee, Junmo Kim

In the recent studies of data augmentation of neural networks, the application of test time augmentation has been studied to extract optimal transformation policies to enhance performance with minimum cost. The policy se…

Data Augmentation

Generalized Wasserstein Dice Loss, Test-time Augmentation, and Transformers for the BraTS 2021 challenge

2021-12-24 · Lucas Fidon, Suprosanna Shit, Ivan Ezhov, Johannes C. Paetzold 외

Brain tumor segmentation from multiple Magnetic Resonance Imaging (MRI) modalities is a challenging task in medical image computation. The main challenges lie in the generalizability to a variety of scanners and imaging …

Brain Tumor SegmentationTumor Segmentation

AugLoss: A Robust Augmentation-based Fine Tuning Methodology

2022-06-05 · Kyle Otstot, Andrew Yang, John Kevin Cava, Lalitha Sankar

Deep Learning (DL) models achieve great successes in many domains. However, DL models increasingly face safety and robustness concerns, including noisy labeling in the training stage and feature distribution shifts in th…

Data Augmentation

SEVA: Leveraging Single-Step Ensemble of Vicinal Augmentations for Test-Time Adaptation

2025-05-07 · Zixuan Hu, Yichun Hu, Ling-Yu Duan

Test-Time adaptation (TTA) aims to enhance model robustness against distribution shifts through rapid model adaptation during inference. While existing TTA methods often rely on entropy-based unsupervised training and ac…

Test-time Adaptation

Feature Augmentation based Test-Time Adaptation

2024-10-18 · Younggeol Cho, Youngrae Kim, Junho Yoon, Seunghoon Hong 외

Test-time adaptation (TTA) allows a model to be adapted to an unseen domain without accessing the source data. Due to the nature of practical environments, TTA has a limited amount of data for adaptation. Recent TTA meth…

Test-time Adaptation