End-to-end learning of CNN features in in discrete optimization models for motion and stereo

Duration: 1 hour 1 min

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/2560946/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

Description:	Pock, T Thursday 7th September 2017 - 09:50 to 10:40


Created:	2017-09-08 08:59
Collection:	Variational methods and effective algorithms for imaging and vision
Publisher:	Isaac Newton Institute
Copyright:	Pock, T
Language:	eng (English)
Distribution:	World (downloadable)
Explicit content:	No
Aspect Ratio:	16:9
Screencast:	No
Bumper:	UCS Default
Trailer:	UCS Default


Abstract:	Co-authors: Patrick Knöbelreiter (Graz University of Technology), Alexander Shekhovtsov (Technical University of Prague), Gottfried Munda (Graz University of Technology), Christian Reinbacher (Amazon) For many years, discrete optimization models such as conditional random fields (CRFs) have defined the state-of-the-art for classical correspondence problems such as motion and stereo. One of the most important ingredients in those models is the choice of the feature transform that is used to compute the similarity between images patches. For a long time, hand crafted features such as the celebrated scale invariant feature transform (SIFT) defined the state-of-the-art. Triggered by the recent success of convolutional neural networks (CNNs), it is quite natural to learn such a feature transform from data. In this talk, I will show how to efficiently learn such CNN features from data using an end-to-end learning approach. It turns out that our learned models yields state-of-the-art results on a number of established benchmark databases. Related Links https://arxiv.org/pdf/1707.06427 - Scalable Full Flow with Learned Binary Descriptors https://arxiv.org/pdf/1611.10229 - End-to-End Training of Hybrid CNN-CRF Models for Stereo

Abstract:

Co-authors: Patrick Knöbelreiter (Graz University of Technology), Alexander Shekhovtsov (Technical University of Prague), Gottfried Munda (Graz University of Technology), Christian Reinbacher (Amazon)

For many years, discrete optimization models such as conditional random fields (CRFs) have defined the state-of-the-art for classical correspondence problems such as motion and stereo. One of the most important ingredients in those models is the choice of the feature transform that is used to compute the similarity between images patches. For a long time, hand crafted features such as the celebrated scale invariant feature transform (SIFT) defined the state-of-the-art. Triggered by the recent success of convolutional neural networks (CNNs), it is quite natural to learn such a feature transform from data. In this talk, I will show how to efficiently learn such CNN features from data using an end-to-end learning approach. It turns out that our learned models yields state-of-the-art results on a number of established benchmark databases.

Related Links
https://arxiv.org/pdf/1707.06427 - Scalable Full Flow with Learned Binary Descriptors
https://arxiv.org/pdf/1611.10229 - End-to-End Training of Hybrid CNN-CRF Models for Stereo

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video	640x360	1.96 Mbits/sec	898.67 MB	View	Download
WebM	640x360	579.32 kbits/sec	258.83 MB	View	Download
iPod Video	480x270	528.21 kbits/sec	236.00 MB	View	Download
MP3	44100 Hz	252.86 kbits/sec	112.98 MB	Listen	Download
Auto *	(Allows browser to choose a format it supports)

Streaming Media Service Upload

End-to-end learning of CNN features in in discrete optimization models for motion and stereo