Near-Duplicated Image Copy Detection based on Orthogonal Fusion of Lightweight Deep Learning Features
Feature vector generation method that can simultaneously preserve global and local information using orthogonal fusion [1], using a lightweight deep learning model MobileNet [2].
[1] Yang et al., "DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features," in Proceedings of the International Conference on Computer Vision, 2021.
[2] Howard et al., "Searching for MobileNetV3," in Proceedings of the International Conference on Computer Vision, 2019.