Scale coding bag of deep features for human attribute and action recognition

• Published in: Machine vision and applications Vol. 29; no. 1; pp. 55 - 71
• Main Authors: Khan, Fahad Shahbaz, van de Weijer, Joost, Anwer, Rao Muhammad, Bagdanov, Andrew D., Felsberg, Michael, Laaksonen, Jorma
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 2018; Springer Nature B.V
• Subjects: Artificial neural networks; Coding; Communications Engineering; Computer Science; Datasets; Feature recognition; Image Processing and Computer Vision; Invariants; Networks; Neural networks; Object recognition; Original Paper; Pattern Recognition; Representations; Vision systems; Willow; Action recognition; Attribute recognition; Bag of deep features
• ISSN: 0932-8092; 1432-1769; 1432-1769
• DOI: 10.1007/s00138-017-0871-1

Most approaches to human attribute and action recognition in still images are based on image representation in which multi-scale local features are pooled across scale into a single, scale-invariant encoding. Both in bag-of-words and the recently popular representations based on convolutional neural networks, local features are computed at multiple scales. However, these multi-scale convolutional features are pooled into a single scale-invariant representation. We argue that entirely scale-invariant image representations are sub-optimal and investigate approaches to scale coding within a bag of deep features framework. Our approach encodes multi-scale information explicitly during the image encoding stage. We propose two strategies to encode multi-scale information explicitly in the final image representation. We validate our two scale coding techniques on five datasets: Willow, PASCAL VOC 2010, PASCAL VOC 2012, Stanford-40 and Human Attributes (HAT-27). On all datasets, the proposed scale coding approaches outperform both the scale-invariant method and the standard deep features of the same network. Further, combining our scale coding approaches with standard deep features leads to consistent improvement over the state of the art.