← Search

Lukas Hedegaard

2 accepted papers

2023

Continual Transformers: Redundancy-Free Attention for Online Inference

ICLR 2023poster

Transformers in their common form are inherently limited to operate on whole token sequences rather than on one token at a time. Consequently, their use during online inference on time-series data entails considerable redundancy due to the overlap in successive token sequences. In this work, we prop…

2022

Continual 3D Convolutional Neural Networks for Real-Time Processing of Videos

ECCV 2022poster

"We introduce Continual 3D Convolutional Neural Networks (Co3D CNNs), a new computational formulation of spatio-temporal 3D CNNs, in which videos are processed frame-by-frame rather than by clip. In online tasks demanding frame-wise predictions, Co3D CNNs dispense with the computational redundancies…