Comparator Networks

Abstract

The objective of this work is set-based verification, e.g. to decide if two sets of images of a face are of the same person or not. The traditional approach to this problem is to learn to generate a feature vector per image, aggregate them into one vector to represent the set, and then compute the cosine similarity between sets. Instead, we design a neural network architecture that can directly learn set-wise verification. Our contributions are: (i) We propose a Deep Comparator Network (DCN) that can ingest a pair of sets (each may contain a variable number of images) as inputs, and compute a similarity between the pair--this involves attending to multiple discriminative local regions (landmarks), and comparing local descriptors between pairs of faces; (ii) To encourage high-quality representations for each set, internal competition is introduced for recalibration based on the landmark score; (iii) Inspired by image retrieval, a novel hard sample mining regime is proposed to control th

Stats

• citations43
• S2 citations—
• github stars0
• HF likes0
• heat score12.33
• arxiv keyxie2018comparator

Related papers

• Conditional Similarity Networks (2016)15.06
• Deep Co-attention Based Comparators For Relative Representation Learning In Person Re-identification (2018)13.34
• Visualizing Deep Similarity Networks (2019)11.85
• Image Similarity Using Deep CNN And Curriculum Learning (2017)0.00
• Correlation Verification For Image Retrieval (2022)18.80
• Relationnet2: Deep Comparison Columns For Few-shot Learning (2018)0.00
• Deepsim-nets: Deep Similarity Networks For Stereo Image Matching (2023)5.24
• Utilizing Complex-valued Network For Learning To Compare Image Patches (2018)0.00