Hiroki Naganuma

Overview

Day1

Workshop on Representation Learning with Very Limited Images

Day2

Workshop on Uncertainty Quantification for Computer Vision

The first work proposes the hypothesis that curvature serves as a key in understanding calibration and OOD generalization. We provide a generalized understanding of Sharpness Aware Minimization, which is effective for generalization, and Focal Loss, which is used in calibration. pic.twitter.com/MIeDawCmin

— Hiroki Naganuma (@_Hiroki11x) October 3, 2023

The second work is an empirical study investigating the impact of pre-trained model selection on downstream tasks in domain generalization. We explore how the size of the pre-trained model and the dataset size influence OOD generalization and calibration.https://t.co/xNcG8v1woG pic.twitter.com/UOYw55yEzZ

— Hiroki Naganuma (@_Hiroki11x) October 3, 2023

Calibrated Out-of-Distribution Detection with a Generic Representation (Oral)

Tomas Vojir, Jan Sochman, Rahaf Aljundi, Jiri Matas

Distance matters for improving performance estimation under covariate shift

Mélanie Roschewitz, Ben Glocker

Gaussian Latent Representations for Uncertainty Estimation using Mahalanobis Distance in Deep Classifiers (Oral)

Aishwarya Venkataramanan, Assia Benbihi, Martin Laviale, Cédric Pradalier

Workshop on OOD Generalization for Computer Vision

Raising the Bar on the Evaluation of Out-of-Distribution Detection[pdf]

Authors: Jishnu Mukhoti (University of Oxford); Tsung-Yu Lin (Facebook AI); Bor-Chun Chen (Facebook AI); Ashish Shah (Facebook AI); Philip Torr (University of Oxford); Puneet Dokania (University of Oxford); Ser-Nam Lim (Meta AI)

Assessing the Impact of Diversity on the Resilience of Deep Learning Ensembles: A Comparative Study on Model Architecture, Output, Activation, and Attribution[pdf]

Authors: Rafael Rosales (Intel); J. Pablo Munoz (Intel); Michael Paulitsch (Intel)

Gradient Estimation for Uneen Domain Risk Minimization with Pre-Trained Models[pdf]

Authors: Byounggyu Lew (Hyperconnect); Donghyun Son (VisualCamp); Buru Chang (Sogang University)

Improving Shift Invariance with Translation Invariant Polyphase Sampling[pdf]

Authors: Sourajit Saha (University of Maryland Baltimore County); Tejas Gokhale (University of Maryland Baltimore County)

Group-Balanced Mixup for Out-of-Distribution Generalization[pdf]

Authors: Sangwoo Hong (Seoul National University); Youngseok Yoon (Seoul National University); Hyungjun Joo (Seoul National University); Jungwoo Lee (Seoul National University)

Weight Averaging Improves Knowledge Distillation under Domain Shift [pdf]

Authors: Valeriy Berezovskiy (HSE University); Nikita Morozov (HSE University)

Mitigating Spurious Correlation in Images by Intervention[pdf]

Authors: Fahimeh HosseiniNoohdani (Sharif university of technology); Mohammad-Mahdi Samiei (Sharif University of Technology); Parsa Hosseini (Sharif University of Technology); Mahdieh Soleymani Baghshah (Sharif University of Technology)

Evaluating Robustness of Pre-Trained Deep Neural Networks against Spurious Correlations [pdf]

Authors: Alireza Hoseinpour (Sharif University of Technology); Majid Taherkhani (Sharif university of technology ); Fahimeh HosseiniNoohdani (Sharif university of technology); Hesam Asadollahzadeh (Sharif University of Technology); Mahdieh Soleymani Baghshah (Sharif University of Technology)

Day3

Day4

Day5

Papers

OOD

DR-Tune: Improving Fine-tuning of Pretrained Visual Models by Distribution Regularization with Semantic Calibration

Nan Zhou, Jiaxin Chen, Di Huang

Understanding the Feature Norm for Out-of-Distribution Detection

Jaewoo Park, Jacky Chen Long Chai, Jaeho Yoon, Andrew Beng Jin Teoh

Learning in Imperfect Environment: Multi-Label Classification with Long-Tailed Distribution and Partial Labels

Wenqiao Zhang, Changshuo Liu, Lingze Zeng, Bengchin Ooi, Siliang Tang, Yueting Zhuang

Nearest Neighbor Guidance for Out-of-Distribution Detection

Jaewoo Park, Yoon Gyo Jung, Andrew Beng Jin Teoh

Distilling Large Vision-Language Model with Out-of-Distribution Generalizability

Xuanlin Li, Yunhao Fang, Minghua Liu, Zhan Ling, Zhuowen Tu, Hao Su

COCO-O: A Benchmark for Object Detectors under Natural Distribution Shifts

Xiaofeng Mao, Yuefeng Chen, Yao Zhu, Da Chen, Hang Su, Rong Zhang, Hui Xue

Anomaly Detection Under Distribution Shift

Tri Cao, Jiawen Zhu, Guansong Pang

Benchmarking Low-Shot Robustness to Natural Distribution Shifts

Aaditya Singh, Kartik Sarangmath, Prithvijit Chattopadhyay, Judy Hoffman

Semi-Supervised Learning via Weight-Aware Distillation under Class Distribution Mismatch

Pan Du, Suyun Zhao, Zisen Sheng, Cuiping Li, Hong Chen

Adaptive Calibrator Ensemble: Navigating Test Set Difficulty in Out-of-Distribution Scenarios

Yuli Zou, Weijian Deng, Liang Zheng

SAFE: Sensitivity-Aware Features for Out-of-Distribution Object Detection

Samuel Wilson, Tobias Fischer, Feras Dayoub, Dimity Miller, Niko Sünderhauf

Calibration

Rethinking Data Distillation: Do Not Overlook Calibration

Dongyao Zhu, Bowen Lei, Jie Zhang, Yanbo Fang, Yiqun Xie, Ruqi Zhang, Dongkuan Xu

When Noisy Labels Meet Long Tail Dilemmas: A Representation Calibration Method

Manyi Zhang, Xuyang Zhao, Jun Yao, Chun Yuan, Weiran Huang

Model Calibration in Dense Classification with Adaptive Label Perturbation

Jiawei Liu, Changkun Ye, Shan Wang, Ruikai Cui, Jing Zhang, Kaihao Zhang, Nick Barnes

RankMixup: Ranking-Based Mixup Training for Network Calibration pdf

Jongyoun Noh, Hyekang Park, Junghyup Lee, Bumsub Ham

Curvature

Enhancing Fine-Tuning Based Backdoor Defense with Sharpness-Aware Minimization pdf

Mingli Zhu, Shaokui Wei, Li Shen, Yanbo Fan, Baoyuan Wu

ImbSAM: A Closer Look at Sharpness-Aware Minimization in Class-Imbalanced Recognition pdf

Yixuan Zhou, Yi Qu, Xing Xu, Hengtao Shen

Curvature-Aware Training for Coordinate Networks pdf

Hemanth Saratchandran, Shin-Fang Chng, Sameera Ramasinghe, Lachlan MacDonald, Simon Lucey

CGBA: Curvature-aware Geometric Black-box Attack pdf

Md Farhamdur Reza, Ali Rahmati, Tianfu Wu, Huaiyu Dai

Understanding Hessian Alignment for Domain Generalization pdf

Sobhan Hemati, Guojun Zhang, Amir Estiri, Xi Chen

Networking

I met a lot of researchers at the UNCV workshop for my poster presentation x 2. The following list is the researchers I met for the first time and remember their names.

Acknowlegements

I want to thank ZOZO Research for supporting my participation in the ICCV. I want to express our deepest gratitude to ZOZO Research for their support.