Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformers

2021-10-07 · Christof Henkel

We present an efficient end-to-end pipeline for largescale landmark recognition and retrieval. We show how to combine and enhance concepts from recent research in image retrieval and introduce two architectures especially suited for large-scale landmark identification. A model with deep orthogonal fusion of local and global features (DOLG) using an EfficientNet backbone as well as a novel Hybrid-Swin-Transformer is discussed and details how to train both architectures efficiently using a step-wise approach and a sub-center arcface loss with dynamic margins are provided. Furthermore, we elaborate a novel discriminative re-ranking methodology for image retrieval. The superiority of our approach was demonstrated by winning the recognition and retrieval track of the Google Landmark Competition 2021.

📄 PDF Abstract BibTeX arXiv:2110.03786

Code (1)

christofhenkel/kaggle-landmark-2021-1st-place 공식 구현 pytorch

Tasks

Image Retrieval Landmark RecognitionRe-RankingRetrieval

Methods 이 논문이 사용한 방법론

Depthwise Convolution Depthwise Convolution is a type of convolution where we apply a single convolutional filter for each input channel. In the regular 2D…

Pointwise Convolution Pointwise Convolution is a type of convolution that uses a 1x1 kernel: a kernel that iterates through every single point. This…

Depthwise Separable Convolution While standard convolution performs the channelwise and spatial-wise computation in one step, Depthwise Separable Convolution …

Average Pooling 설명 없음

1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…

Batch Normalization 설명 없음

Sigmoid Activation 설명 없음

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Computationally Bounded Retrieval

2015-06-01 · CVPR 2015 6 · Mohammad Rastegari, Cem Keskin, Pushmeet Kohli, Shahram Izadi

The increase in size of large image databases makes the problem of efficient retrieval extremely challenging. This is especially true in the case of high dimensional data where even operations like hashing become expensi…

Attribute Image Retrieval Retrieval

Orthogonal Projection Loss

2021-03-25 · ICCV 2021 10 · Kanchana Ranasinghe, Muzammal Naseer, Munawar Hayat, Salman Khan 외

Deep neural networks have achieved remarkable performance on a range of classification tasks, with softmax cross-entropy (CE) loss emerging as the de-facto objective function. The CE loss encourages features of a class t…

Domain Generalization Few-Shot Learning

SOFAR: large-scale association network learning

2017-04-26 · Yoshimasa Uematsu, Yingying Fan, Kun Chen, Jinchi Lv 외

Many modern big data applications feature large scale in both numbers of responses and predictors. Better statistical efficiency and scientific insights can be enabled by understanding the large-scale response-predictor …

Large-Scale Image Retrieval with Attentive Deep Local Features

2016-12-19 · ICCV 2017 10 · Hyeonwoo Noh, Andre Araujo, Jack Sim, Tobias Weyand 외

We propose an attentive local feature descriptor suitable for large-scale image retrieval, referred to as DELF (DEep Local Feature). The new feature is based on convolutional neural networks, which are trained only with …

Image Retrieval Retrieval

Survey on Sparse Coded Features for Content Based Face Image Retrieval

2014-02-20 · D. Johnvictor, G. Selvavinayagam

Content based image retrieval, a technique which uses visual contents of image to search images from large scale image databases according to users' interests. This paper provides a comprehensive survey on recent technol…

Content-Based Image Retrieval Face Image Retrieval Image Retrieval Retrieval