Team Name: Smart Image

Team Member: Lingxi Xie, Xiaopeng Zhang, Bingcheng Liu, Zhao Yang, Zewei Du, Hang Chen, Longhui Wei, Yaxiong Chi

Team Affiliation: Huawei Cloud & Huawei 2012 Labs

Method Description (short) : Our work is implemented on Huawei ModelArts platform [1], which slightly improves accuracy while being much faster in training. As for the algorithms, the main idea is to leverage area under the margin and knowledge distillation for handling noise labels, as well as an algorithm for learning an ensemble model.

Team Name: fISHpAM

Team Member: Canxiang Yan, Cheng Niu, Jie Zhou

Team Affiliation: Pattern Recognition Center, WeChat AI, Tencent Inc, China.

Method Description (short) : We use pretraining and ensembling techniques to improve the performance. Using WordNet, each image can be mapped to several word tags (e.g., noun and adjective.). Then base models are pretrained with those multi-label images and different network archtectures.

Team Name: PCI-AI

Team Member: Zhiwei Wu, Shuwen Sun, Kunmin Li, Rui Zhang, Zhenjie Huang, Yanyi Feng

Team Affiliation: pcitech (https://www.pcitech.com/)

Method Description (short) : Our method is based on the ResNet and ResNet variants, ResNet101,ResNet152[1], ResNext101[2] and ResNest101[3].

Paper Title: Smooth Proxy-Anchor Loss for Noisy Metric Learning

Authors: Carlos Roig, David Varas, Issey Masuda, Juan Carlos Riveiro, Elisenda Bou-Balust

Email: {carlos,david.varas,issey,eli}@vilynx.com

Short Description: Smooth Proxy-Anchor Loss for Noisy Metric Learning tackles the problem of having noisy samples in a metric learning problem with a novel architecture.

Keywords: Metric Learning, Noisy data.

Paper Title: OSVGAN: Generative Adversarial Networks for Data Scarce Online Signature Verification

Authors: Chandra Sekhar Vorugunti, Sai Sasikanth Indukuri, Viswanath Pulabaigari and Rama Krishna Sai Gorthi

Email: Chandrasekhar.v@iiits.in, sindukuri@umass.edu, viswanath.p@iiits.ac.in, rkg@iittp.ac.in

Short Description: In this work, two most challenging requirements of Online Signature Verification (OSV) are addressed. First, data scarcity to thoroughly test the framework for real time deployment in critical applications. Second, achieving few shot learning, especially one-shot learning to classify the genuineness of test signature with as minimum as one training sample per user.

Keywords: Synthetic Online Signature generation, Generative Adversarial Networks, Online Signature verification.

Paper Title: When Ensembling Smaller Models is More Efficient than Single Large Models

Authors: Dan Kondratyuk, Mingxing Tan, Matthew Brown, Boqing Gong

Email: {dankondratyuk,tanmingxing,mtbr,bgong}@google.com

Short Description: Ensembling is a simple and popular technique for boosting evaluation performance by training multiple models. This approach is commonly reserved for the largest models, as it is commonly held that increasing the model size provides a more substantial reduction in error than ensembling smaller models. However, we show results from experiments on CIFAR-10 and ImageNet that ensembles can outperform single models with both higher accuracy and requiring fewer total FLOPs to compute. This can imply output diversity in ensembling can often be more efficient than training larger models, especially when the models approach the size of the dataset.

Keywords: ensemble, efficient, NAS, vision.

Paper Title: Distilling Effective Supervision From Severe Label Noise

Authors: Zizhao Zhang, Han Zhang, Sercan Ö. Arık, Honglak Lee, Tomas Pfister

Email: zizhaoz@google.com, zhanghan@google.com, soarik@google.com, honglak@google.com, tpfister@google.com

Short Description: We estimate Data Coefficients with a generalized meta learning framework and set new state of the arts on noisy label benchmarks. Code is available.

Keywords: Robust training, Meta learning, Noise labels.

Paper Title: Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation

Authors: Min-Hung Chen, Baopu Li, Yingze Bao, Ghassan AlRegib, Zsolt Kira

Email: cmhungsteve@gatech.edu, baopuli@baidu.com, baoyingze@baidu.com, alregib@gatech.edu, zkira@gatech.edu

Short Description: A method for cross-domain action segmentation by aligning feature spaces across multiple temporal scales to reduce spatio-temporal variability.

Keywords: domain adaptation, action segmentation, self-supervised learning, video understanding, temporal dynamics, domain discrepancy, temporal variations, multi-scale.

Paper Title: Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition From a Domain Adaptation Perspective

Authors: Muhammad Abdullah Jamal, Matthew Brown, Ming-Hsuan Yang, Liqiang Wang, Boqing Gong

Email: Muhammad Abdullah Jamal, Matthew Brown, Ming-Hsuan Yang, Liqiang Wang, Boqing Gong, bgong@google.com

Short Description: In this paper, we analyze the long-tail problem from domain adaptation perspective. We propose a reweighting approach and validate it on six datasets.

Keywords: Long-tail Visual Recognition, Domain Adaptation, Meta-Learning, Class-imbalance.