UberNet: Training a `Universal' Convolutional Neural Network for Low-, Mid-, and High-Level Vision using Diverse Datasets and Limited Memory
This work provides a unified solution for various vision tasks, which is incremental as it builds on existing CNN methods to handle multiple tasks jointly.
The paper tackles the problem of training a single convolutional neural network to handle multiple vision tasks (low-, mid-, and high-level) by addressing challenges in diverse datasets and limited memory, achieving competitive performance across seven tasks in 0.7 seconds per frame on a GPU.
In this work we introduce a convolutional neural network (CNN) that jointly handles low-, mid-, and high-level vision tasks in a unified architecture that is trained end-to-end. Such a universal network can act like a `swiss knife' for vision tasks; we call this architecture an UberNet to indicate its overarching nature. We address two main technical challenges that emerge when broadening up the range of tasks handled by a single CNN: (i) training a deep architecture while relying on diverse training sets and (ii) training many (potentially unlimited) tasks with a limited memory budget. Properly addressing these two problems allows us to train accurate predictors for a host of tasks, without compromising accuracy. Through these advances we train in an end-to-end manner a CNN that simultaneously addresses (a) boundary detection (b) normal estimation (c) saliency estimation (d) semantic segmentation (e) human part segmentation (f) semantic boundary detection, (g) region proposal generation and object detection. We obtain competitive performance while jointly addressing all of these tasks in 0.7 seconds per frame on a single GPU. A demonstration of this system can be found at http://cvn.ecp.fr/ubernet/.