paper-with-me

홈 › Papers

Results and findings of the 2021 Image Similarity Challenge

2022-02-08 · Zoë Papakipos, Giorgos Tolias, Tomas Jenicek, Ed Pizzi, Shuhei Yokoo, Wenhao Wang, Yifan Sun, Weipu Zhang, Yi Yang, Sanjay Addicam, Sergio Manuel Papadakis, Cristian Canton Ferrer, Ondrej Chum, Matthijs Douze

The 2021 Image Similarity Challenge introduced a dataset to serve as a new benchmark to evaluate recent image copy detection methods. There were 200 participants to the competition. This paper presents a quantitative and qualitative analysis of the top submissions. It appears that the most difficult image transformations involve either severe image crops or hiding into unrelated images, combined with local pixel perturbations. The key algorithmic elements in the winning submissions are: training on strong augmentations, self-supervised learning, score normalization, explicit overlay detection, and global descriptor matching followed by pairwise image comparison.

📄 PDF Abstract BibTeX arXiv:2202.04007

Code (0)

등록된 구현이 없습니다.

Tasks

Copy DetectionSelf-Supervised Learning

Similar Papers 제목 키워드 기반

Quantifying the Effect of Image Similarity on Diabetic Foot Ulcer Classification

2023-04-25 · Imran Chowdhury Dipto, Bill Cassidy, Connah Kendrick, Neil D. Reeves 외

This research conducts an investigation on the effect of visually similar images within a publicly available diabetic foot ulcer dataset when training deep learning classification networks. The presence of binary-identic…

Deep Learning

Efficient Discovery and Effective Evaluation of Visual Perceptual Similarity: A Benchmark and Beyond

2023-08-28 · ICCV 2023 1 · Oren Barkan, Tal Reiss, Jonathan Weill, Ori Katz 외

Visual similarities discovery (VSD) is an important task with broad e-commerce applications. Given an image of a certain object, the goal of VSD is to retrieve images of different objects with high perceptual visual simi…

Retrieval

EMBEDDIA at SemEval-2022 Task 8: Investigating Sentence, Image, and Knowledge Graph Representations for Multilingual News Article Similarity

2022-07-01 · SemEval (NAACL) 2022 7 · Elaine Zosa, Emanuela Boros, Boshko Koloski, Lidia Pivovarova

In this paper, we present the participation of the EMBEDDIA team in the SemEval-2022 Task 8 (Multilingual News Article Similarity). We cover several techniques and propose different methods for finding the multilingual n…

ArticlesSemantic SimilaritySemantic Textual SimilaritySentence

Magnitude Matters: a Superior Class of Similarity Metrics for Holistic Semantic Understanding

2025-09-12 · V. S. Raghu Parupudi arxiv

Vector comparison in high dimensions is a fundamental task in NLP, yet it is dominated by two baselines: the raw dot product, which is unbounded and sensitive to vector norms, and the cosine similarity, which discards ma…

Mapping fMRI Signal and Image Stimuli in an Artificial Neural Network Latent Space: Bringing Artificial and Natural Minds Together

2025-03-12 · Cesare Maria Dalbagno, Manuel de Castro Ribeiro Jardim, Mihnea Angheluţă

The goal of this study is to investigate whether latent space representations of visual stimuli and fMRI data share common information. Decoding and reconstructing stimuli from fMRI data remains a challenge in AI and neu…