Timezone: »

 
Poster
Revisiting Multi-Task Learning with ROCK: a Deep Residual Auxiliary Block for Visual Detection
Taylor Mordan · Nicolas THOME · Gilles Henaff · Matthieu Cord

Tue Dec 04 02:00 PM -- 04:00 PM (PST) @ Room 210 #60
in Tue Poster Session B »

Multi-Task Learning (MTL) is appealing for deep learning regularization. In this paper, we tackle a specific MTL context denoted as primary MTL, where the ultimate goal is to improve the performance of a given primary task by leveraging several other auxiliary tasks. Our main methodological contribution is to introduce ROCK, a new generic multi-modal fusion block for deep learning tailored to the primary MTL context. ROCK architecture is based on a residual connection, which makes forward prediction explicitly impacted by the intermediate auxiliary representations. The auxiliary predictor's architecture is also specifically designed to our primary MTL context, by incorporating intensive pooling operators for maximizing complementarity of intermediate representations. Extensive experiments on NYUv2 dataset (object detection with scene classification, depth prediction, and surface normal estimation as auxiliary tasks) validate the relevance of the approach and its superiority to flat MTL approaches. Our method outperforms state-of-the-art object detection models on NYUv2 dataset by a large margin, and is also able to handle large-scale heterogeneous inputs (real and synthetic images) with missing annotation modalities.

Keywords: Computer Vision · Multitask and Transfer Learning · Object Detection · CNN Architectures
[ Paper] 

Author Information

Taylor Mordan (Sorbonne Université, LIP6)
Nicolas THOME (Cnam)
Gilles Henaff (Thales Optronique S.A.S.)
Matthieu Cord (Sorbonne University)

More from the Same Authors