A Saccaded Visual Transformer for General Object Spotting

Autor:	Pye, Willem. T., Sinclair, David. A.
Rok vydání:	2022
Předmět:	Computer Science - Computer Vision and Pattern Recognition
Druh dokumentu:	Working Paper
Popis:	This paper presents the novel combination of a visual transformer style patch classifier with saccaded local attention. A novel optimisation paradigm for training object models is also presented, rather than the optimisation function minimising class membership probability error the network is trained to estimate the normalised distance to the centroid of labelled objects. This approach builds a degree of transnational invariance directly into the model and allows fast saccaded search with gradient ascent to find object centroids. The resulting saccaded visual transformer is demonstrated on human faces. Comment: 11 pages mostly figure, central idea is to train on distance a patch is form a labelled feature
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2210.09220 Zobrazit plný text záznamu View this record from Arxiv