Learning Phase-Based Descriptions for Action Recognition

Action recognition continues to receive significant attention from the research community, with new neural network architectures being developed continuously. Optical flow is by far the most popular input motion representation to these architectures, leaving a lot of undiscovered potential for other types of motion representations. Eulerian...

Top-Down Networks: A coarse-to-fine reimagination of CNNs

Biological vision adopts a coarse-to-fine information processing pathway, from initial visual detection and binding of salient features of a visual scene, to the enhanced and preferential processing given relevant stimuli. On the contrary, CNNs employ a fine-to-coarse processing, moving from local, edge-detecting filters to more global ones...

VideoMuscle: Estimating human musculoskeletal model using neural networks from monocular video

We present the first deep learning approach to estimate the human skeletal system of the musculoskeletal model from monocular video. The current practice of musculoskeletal modeling relies on a motion capture system and OpenSim. The data is recorded in a restricted environment, and OpenSim workflow for musculoskeletal modeling is costly. Our...

Scale Learning in Scale-Equivariant Convolutional networks

In real-life scenarios, there are many variations in sizes of objects of the same category and the objects are not always placed at a fixed distance from the camera. This results in objects taking up an arbitrary size of pixels in the image. Vanilla CNNs are by design only translation equivariant and thus have to learn separate filters for...

Rotation invariant filters in CNNs: applied to segmentation of aerial images for land-use classification

Convolutional neural networks are showing incredible performance in image classification, segmentation, object detection and other computer vision applications in recent years. But they lack understanding of affine transformations to input data. In this work, we introduce rotational invariant<br/>convolutional neural networks that learn...

Data-efficient resolution transfer with Continuous Kernel Convolutional Neural Networks

Convolutional Neural Networks (CNNs) benefit from fine-grained details in high-resolution images, but these images are not always easily available as data collection can be expensive or time-consuming. Transfer learning pre-trains models on data from a related domain before fine-tuning on the main domain, and is a common strategy to deal with...