Colorization as a Proxy Task for Visual Understanding
Gustav Larsson, Michael Maire, Gregory Shakhnarovich · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · 2017
We investigate and improve self-supervision as a drop-in replacement for ImageNet pretraining, focusing on automatic colorization as the proxy task. Self-supervised training has been shown to be more promising for utilizing unlabeled data than other, traditional unsupervised learning methods. We build on this success and evaluate the ability of our self-supervised network in several contexts. On VOC segmentation and classification tasks, we present results that are state-of-the-art among methods not using ImageNet labels for pretraining representations. Moreover, we present the first in-depth analysis of self-supervision via colorization, concluding that formulation of the loss, training details and network architecture play important roles in its effectiveness. This investigation is further expanded by revisiting the ImageNet pretraining paradigm, asking questions such as: How much training data is needed? How many labels are needed? How much do features change when fine-tuned? We relate these questions back to self-supervision by showing that colorization provides a similarly powerful supervisory signal as various flavors of ImageNet pretraining.
Cited by 458 works. doi:10.1109/cvpr.2017.96
Authors
References
- Adam: A Method for Stochastic Optimization
- ImageNet Large Scale Visual Recognition Challenge
- Fully convolutional networks for semantic segmentation
- AutoResearch Lite: a reproducible paper-inspired experiment runner (Adam mechanism reproduction and convergence-speed comparison)
- Caffe
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Rectified Linear Units Improve Restricted Boltzmann Machines
- Understanding the difficulty of training deep feedforward neural networks
- Caffe: Convolutional Architecture for Fast Feature Embedding