Thesis
Self-supervised learning using motion and visualizing convolutional neural networks
- Abstract:
-
Expand abstract
We propose a novel method for learning convolutional image representations without manual supervision. We use motion in the form of optical-flow, to supervise representations of static images. Training a network to predict flow from a single image can be needlessly difficult due to intrinsic ambiguities in this prediction task. We instead propose two simpler learning goals: (a) embed pixels such that the similarity between their embeddings matches that between their optical-flow vectors (C...
Actions
Authors
Contributors
Funding
Bibliographic Details
- DOI:
- Type of award:
- DPhil
- Level of award:
- Doctoral
- Awarding institution:
- University of Oxford
Item Description
- Language:
-
English
- Keywords:
- Subjects:
- UUID:
-
uuid:05ef7004-0bb1-4852-be1f-892daf694430
- Deposit date:
-
2019-02-23
Related Items
Image representations, from SIFT and bag of visual words to Convolutional Neural Networks (CNNs) are a crucial component of almost all computer vision systems. However, our understanding of them remains limited. In this paper we study several landmark representations, both shallow and deep, by a number of complementary visualization techniques. These visualizations are based on the concept of “natural pre-image”, namely a natural-looking image whose representation has some notable property. ...
We propose a novel method for learning convolutional neural image representations without manual supervision. We use motion cues in the form of optical-flow, to supervise representations of static images. The obvious approach of training a network to predict flow from a single image can be needlessly difficult due to intrinsic ambiguities in this prediction task. We instead propose a much simpler learning goal: embed pixels such that the similarity between their embeddings matches that betwee...
Deconvolutional networks are a popular architecture. However, when used to reverse an existing model for visualization purposes, the process involves a few heuristic choices that cannot be immediately interpreted. In this paper, we introduce a family of reversed networks which include deconvolution, backpropagation and network saliency as special but important cases. We use this construction to thoroughly investigate the different architectures and compare them in terms of quality of the prod...