Over the last years deep learning methods have been shown to outperform previous state‐of‐
the‐art machine learning techniques in several fields, with computer vision being one of the …

Recognizing human activities from video sequences or still images is a challenging task due
to problems, such as background clutter, partial occlusion, changes in scale, viewpoint …

On-device training enables the model to adapt to new data collected from the sensors by
fine-tuning a pre-trained model. Users can benefit from customized AI models without having …

Existing state-of-the-art saliency detection methods heavily rely on CNN-based
architectures. Alternatively, we rethink this task from a convolution-free sequence-to …

Human activity recognition (HAR) systems attempt to automatically identify and analyze
human activities using acquired information from various types of sensors. Although several …

Most state-of-the-art action localization systems process each action proposal individually,
without explicitly exploiting their relations during learning. However, the relations between …

The explosive growth in video streaming gives rise to challenges on performing video
understanding at high accuracy and low computation cost. Conventional 2D CNNs are …

Abstract Unsupervised Domain Adaptive (UDA) person re-identification (ReID) aims at
adapting the model trained on a labeled source-domain dataset to a target-domain dataset …

Efficient on-device learning requires a small memory footprint at training time to fit the tight
memory constraint. Existing work solves this problem by reducing the number of trainable …

We propose a technique for producing'visual explanations' for decisions from a large class
of Convolutional Neural Network (CNN)-based models, making them more transparent. Our …