paper-with-me

Papers

Learning to Transform for Generalizable Instance-wise Invariance

2023-09-28 · ICCV 2023 1 · Utkarsh Singhal, Carlos Esteves, Ameesh Makadia, Stella X. Yu

Computer vision research has long aimed to build systems that are robust to spatial transformations found in natural data. Traditionally, this is done using data augmentation or hard-coding invariances into the architecture. However, too much or too little invariance can hurt, and the correct amount is unknown a priori and dependent on the instance. Ideally, the appropriate invariance would be learned from data and inferred at test-time. We treat invariance as a prediction problem. Given any image, we use a normalizing flow to predict a distribution over transformations and average the predictions over them. Since this distribution only depends on the instance, we can align instances before classifying them and generalize invariance across classes. The same distribution can also be used to adapt to out-of-distribution poses. This normalizing flow is trained end-to-end and can learn a much larger range of transformations than Augerino and InstaAug. When used as data augmentation, our method shows accuracy and robustness gains on CIFAR 10, CIFAR10-LT, and TinyImageNet.

📄 PDF Abstract BibTeX arXiv:2309.16672

Code (1)

sutkarsh/flow_inv 공식 구현 pytorch

Tasks

Data Augmentation

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Scale-invariant unconstrained online learning

2017-08-23 · Wojciech Kotłowski

We consider a variant of online convex optimization in which both the instances (input vectors) and the comparator (weight vector) are unconstrained. We exploit a natural scale invariance symmetry in our unconstrained se…

PointClustering: Unsupervised Point Cloud Pre-Training Using Transformation Invariance in Clustering

2023-01-01 · CVPR 2023 1 · Fuchen Long, Ting Yao, Zhaofan Qiu, Lusong Li 외

Feature invariance under different data transformations, i.e., transformation invariance, can be regarded as a type of self-supervision for representation learning. In this paper, we present PointClustering, a new un…

ClusteringDeep ClusteringInductive BiasPhilosophy+2

Volumetric Transformer Networks

2020-07-18 · ECCV 2020 8 · Seungryong Kim, Sabine Süsstrunk, Mathieu Salzmann

Existing techniques to encode spatial invariance within deep convolutional neural networks (CNNs) apply the same warping field to all the feature channels. This does not account for the fact that the individual feature c…

DecoderFine-Grained Image RecognitionImage RetrievalRetrieval

Group-based Learning of Disentangled Representations with Generalizability for Novel Contents

2018-09-07 · Haruo Hosoya

Sensory data are often comprised of independent content and transformation factors. For example, face images may have shapes as content and poses as transformation. To infer separately these factors from given data, vari…

Attribute

Consistency as Inductive Bias: Learning Cross-View Invariance for Robust Multimodal Reasoning

2026-06-29 · Xin Zou, Haolin Deng, Yibo Yan, Shuliang Liu 외 arxiv

Inductive biases steer learning toward generalizable solutions by encoding task structure. In this work, we identify a crucial missing bias in MLLMs: cross-view consistency, \textit{i.e.}, semantically invariant views of…

Reinforcement LearningMultimodal ReasoningData Augmentation