Unsupervised Monocular Depth Estimation with Left-Right Consistency
- Clément Godard, Oisin Mac Aodha, G. Brostow
- Computer ScienceComputer Vision and Pattern Recognition
- 13 September 2016
This paper proposes a novel training objective that enables the convolutional neural network to learn to perform single image depth estimation, despite the absence of ground truth depth data, and produces state of the art results for monocular depth estimation on the KITTI driving dataset.
Digging Into Self-Supervised Monocular Depth Estimation
- Clément Godard, Oisin Mac Aodha, G. Brostow
- Computer ScienceIEEE International Conference on Computer Vision
- 4 June 2018
It is shown that a surprisingly simple model, and associated design choices, lead to superior predictions, and together result in both quantitatively and qualitatively improved depth maps compared to competing self-supervised methods.
Semantic object classes in video: A high-definition ground truth database
- G. Brostow, J. Fauqueur, R. Cipolla
- Computer SciencePattern Recognition Letters
- 15 January 2009
Segmentation and Recognition Using Structure from Motion Point Clouds
- G. Brostow, J. Shotton, J. Fauqueur, R. Cipolla
- Computer ScienceEuropean Conference on Computer Vision
- 20 October 2008
This work proposes an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion that works well on sparse, noisy point clouds, and unlike existing approaches, does not need appearance-based descriptors.
Harmonic Networks: Deep Translation and Rotation Equivariance
- Daniel E. Worrall, Stephan J. Garbin, Daniyar Turmukhambetov, G. Brostow
- Computer ScienceComputer Vision and Pattern Recognition
- 14 December 2016
H-Nets are presented, a CNN exhibiting equivariance to patch-wise translation and 360-rotation, and it is demonstrated that their layers are general enough to be used in conjunction with the latest architectures and techniques, such as deep supervision and batch normalization.
Self-Supervised Monocular Depth Hints
- Jamie Watson, Michael Firman, G. Brostow, Daniyar Turmukhambetov
- GeologyIEEE International Conference on Computer Vision
- 19 September 2019
This work studies the problem of ambiguous reprojections in depth-prediction from stereo-based self-supervision, and introduces Depth Hints to alleviate their effects, and produces state-of-the-art depth predictions on the KITTI benchmark.
Patch Based Synthesis for Single Depth Image Super-Resolution
- Oisin Mac Aodha, N. Campbell, A. Nair, G. Brostow
- Computer ScienceEuropean Conference on Computer Vision
- 7 October 2012
This work presents an algorithm to synthetically increase the resolution of a solitary depth image using only a generic database of local patches, and shows how important further depth-specific processing, such as noise removal and correct patch normalization, dramatically improves results.
Unsupervised Bayesian Detection of Independent Motion in Crowds
- G. Brostow, R. Cipolla
- Computer ScienceComputer Vision and Pattern Recognition
- 17 June 2006
An unsupervised data driven Bayesian clustering algorithm which has detection of individual entities as its primary goal and can be augmented with subject-specific filtering, but is shown to already be effective at detecting individual entities in crowds of people, insects, and animals.
Structured Prediction of Unobserved Voxels from a Single Depth Image
- Michael Firman, Oisin Mac Aodha, S. Julier, G. Brostow
- Computer ScienceComputer Vision and Pattern Recognition
- 1 June 2016
This work proposes an algorithm that can complete the unobserved geometry of tabletop-sized objects, based on a supervised model trained on already available volumetric elements, that maps from a local observation in a single depth image to an estimate of the surface shape in the surrounding neighborhood.
The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth
- Jamie Watson, Oisin Mac Aodha, V. Prisacariu, G. Brostow, Michael Firman
- Computer ScienceComputer Vision and Pattern Recognition
- 29 April 2021
ManyDepth is proposed, an adaptive approach to dense depth estimation that can make use of sequence information at test time, when it is available, and takes inspiration from multi-view stereo, a deep end-to-end cost volume based approach that is trained using self-supervision only.
...
...