Dynamic neural network is an emerging research topic in deep learning. Compared to static
models which have fixed computational graphs and parameters at the inference stage …

This paper presents X3D, a family of efficient video networks that progressively expand a
tiny 2D image classification architecture along multiple network axes, in space, time, width …

The performance of video prediction has been greatly boosted by advanced deep neural
networks. However, most of the current methods suffer from large model sizes and require …

Data redundancy is ubiquitous in the inputs and intermediate results of Deep Neural
Networks (DNN). It offers many significant opportunities for improving DNN performance and …

Very deep convolutional neural networks offer excellent recognition results, yet their
computational expense limits their impact for many real-world applications. We introduce …

In the face of the video data deluge, today's expensive clip-level classifiers are increasingly
impractical. We propose a framework for efficient action recognition in untrimmed video that …

Our paper presents a new approach for temporal detection of human actions in long,
untrimmed video sequences. We introduce Single-Stream Temporal Action Proposals (SST) …

We present AdaFrame, a framework that adaptively selects relevant frames on a per-input
basis for fast video recognition. AdaFrame contains a Long Short-Term Memory network …

Recurrent Neural Networks (RNNs) continue to show outstanding performance in sequence
modeling tasks. However, training RNNs on long sequences often face challenges like slow …

Adaptive optimization methods for deep learning adjust the inference task to the current
circumstances at runtime to improve the resource footprint while maintaining the model's …