no code implementations • 13 Feb 2024 • Rasmus Kjær Høier, Christopher Zach
The search for "biologically plausible" learning algorithms has converged on the idea of representing gradients as activity differences.
no code implementations • 4 Apr 2023 • Christopher Zach
By using the underlying theory of proper scoring rules, we design a family of noise-contrastive estimation (NCE) methods that are tractable for latent variable models.
1 code implementation • 2 Feb 2023 • Rasmus Høier, D. Staudt, Christopher Zach
Activity difference based learning algorithms-such as contrastive Hebbian learning and equilibrium propagation-have been proposed as biologically plausible alternatives to error back-propagation.
1 code implementation • CVPR 2023 • Xixi Liu, Yaroslava Lochman, Christopher Zach
Its performance is demonstrated on the large-scale ImageNet-1k OOD detection benchmark.
1 code implementation • 1 Apr 2022 • PoHao Hsu, Che-Tsung Lin, Chun Chet Ng, Jie-Long Kew, Mei Yih Tan, Shang-Hong Lai, Chee Seng Chan, Christopher Zach
Deep learning-based methods have made impressive progress in enhancing extremely low-light images - the image quality of the reconstructed images has generally improved.
no code implementations • CVPR 2022 • Huu Le, Rasmus Kjær Høier, Che-Tsung Lin, Christopher Zach
We propose a new algorithm for training deep neural networks (DNNs) with binary weights.
1 code implementation • ICCV 2021 • Yaroslava Lochman, Kostiantyn Liepieshov, Jianhui Chen, Michal Perdoch, Christopher Zach, James Pritts
A lot of the difficulties of general camera calibration lie in the use of a forward projection model.
no code implementations • 15 May 2021 • Christopher Zach
In this work we unify a number of inference learning methods, that are proposed in the literature as alternative training algorithms to the ones based on regular error back-propagation.
no code implementations • 22 Feb 2021 • Huu Le, Christopher Zach
Robust parameter estimation is a crucial task in several 3D computer vision pipelines such as Structure from Motion (SfM).
1 code implementation • 21 Oct 2020 • Huu Le, Christopher Zach, Edward Rosten, Oliver J. Woodford
Non-linear least squares solvers are used across a broad range of offline and real-time model fitting problems.
no code implementations • 7 May 2020 • Rasmus Kjær Høier, Christopher Zach
In this work we propose lifted regression/reconstruction networks (LRRNs), which combine lifted neural networks with a guaranteed Lipschitz continuity property for the output layer.
1 code implementation • CVPR 2020 • Huu Le, Christopher Zach
Due to the highly non-convex nature of large-scale robust parameter estimation, avoiding poor local minima is challenging in real-world applications where input data is contaminated by a large or unknown fraction of outliers.
no code implementations • ECCV 2020 • Christopher Zach, Huu Le
Optimization problems with an auxiliary latent variable structure in addition to the main model parameters occur frequently in computer vision and machine learning.
no code implementations • ECCV 2020 • Pulak Purkait, Christopher Zach, Ian Reid
Our method learns the co-occurrences, and appearance parameters such as shape and pose, for different objects categories through a grammar-based auto-encoder, resulting in a compact and accurate representation for scene layouts.
no code implementations • ICCV 2019 • Christopher Zach, Guillaume Bourmaud
Robust cost optimization is the task of fitting parameters to data points containing outliers.
no code implementations • 19 Jun 2019 • Eskil Jörgensen, Christopher Zach, Fredrik Kahl
We show how modeling heteroscedastic uncertainty improves performance upon our baseline, and furthermore, how back-propagation can be done through the optimizer in order to train the pipeline end-to-end for additional accuracy.
no code implementations • 7 Jun 2019 • Pulak Purkait, Christopher Zach, Ian Reid
In our experiments we demonstrate that a CNN trained by minimizing the proposed loss is able to predict semantic categories for visible and occluded object parts without requiring to increase the network size (compared to a standard segmentation task).
no code implementations • 7 May 2019 • Christopher Zach, Virginia Estellers
In this work we address supervised learning of neural networks via lifted network formulations.
no code implementations • ECCV 2018 • Christopher Zach, Guillaume Bourmaud
Robust cost optimization is the challenging task of fitting a large number of parameters to data points containing a significant and unknown fraction of outliers.
1 code implementation • 10 Aug 2018 • Pulak Purkait, Ujwal Bonde, Christopher Zach
A major element of depth perception and 3D understanding is the ability to predict the 3D layout of a scene and its contained objects for a novel pose.
no code implementations • CVPR 2018 • Je Hyeong Hong, Christopher Zach
Bundle adjustment is a nonlinear refinement method for camera poses and 3D structure requiring sufficiently good initialization.
2 code implementations • 11 May 2018 • Rudra P. K. Poudel, Ujwal Bonde, Stephan Liwicki, Christopher Zach
Modern deep learning architectures produce highly accurate results on many challenging semantic segmentation datasets.
Ranked #79 on Semantic Segmentation on Cityscapes val
no code implementations • 22 Mar 2018 • Pulak Purkait, Christopher Zach, Anders Eriksson
Robust parameter estimation in computer vision is frequently accomplished by solving the maximum consensus (MaxCon) problem.
1 code implementation • 9 Dec 2017 • Pulak Purkait, Cheng Zhao, Christopher Zach
In this work we design a deep neural network architecture based on sparse feature descriptors to estimate the absolute pose of an image.
no code implementations • 8 Dec 2017 • Pulak Purkait, Christopher Zach
Modern automotive vehicles are often equipped with a budget commercial rolling shutter camera.
no code implementations • ICCV 2017 • Pulak Purkait, Christopher Zach, Ales Leonardis
A vast majority of consumer cameras operate the rolling shutter mechanism, which often produces distorted images due to inter-row delay while capturing an image.
no code implementations • CVPR 2017 • Je Hyeong Hong, Christopher Zach, Andrew Fitzgibbon
Variable Projection (VarPro) is a framework to solve optimization problems efficiently by optimally eliminating a subset of the unknowns.
no code implementations • ICCV 2015 • Andrea Cohen, Christopher Zach
Robust estimation of model parameters in the presence of outliers is a key problem in computer vision.
no code implementations • CVPR 2015 • Christopher Zach, Adrian Penate-Sanchez, Minh-Tri Pham
Joint object recognition and pose estimation solely from range images is an important task e. g. in robotics applications and in automated manufacturing environments.
no code implementations • 12 Nov 2014 • Tian Cao, Christopher Zach, Shannon Modla, Debbie Powell, Kirk Czymmek, Marc Niethammer
In this report, I introduce two methods of image registration for correlative microscopy.
no code implementations • CVPR 2014 • Christopher Zach
In this work we reconsider labeling problems with (virtually) continuous state spaces, which are of relevance in low level computer vision.
no code implementations • 14 Aug 2013 • Christopher Zach, Christian Häne
The number of unknowns is O(LK) per pairwise clique in terms of the state space size $L$ and the number of linear segments K. This compares to an O(L^2) size complexity of the standard LP relaxation if the piecewise linear structure is ignored.
no code implementations • CVPR 2013 • Jamie Shotton, Ben Glocker, Christopher Zach, Shahram Izadi, Antonio Criminisi, Andrew Fitzgibbon
We address the problem of inferring the pose of an RGB-D camera relative to a known 3D scene, given only a single acquired image.
no code implementations • CVPR 2013 • Christian Hane, Christopher Zach, Andrea Cohen, Roland Angst, Marc Pollefeys
Image segmentations provide geometric cues about which surface orientations are more likely to appear at a certain location in space whereas a dense 3D reconstruction yields a suitable regularization for the segmentation problem by lifting the labeling from 2D images to 3D space.
no code implementations • NeurIPS 2010 • Roland Memisevic, Christopher Zach, Marc Pollefeys, Geoffrey E. Hinton
We describe a log-bilinear" model that computes class probabilities by combining an input vector multiplicatively with a vector of binary latent variables.