Zobrazeno 1 - 10
of 51
pro vyhledávání: '"Sullivan, Josephine"'
We construct a strong baseline method for building damage detection by starting with the highly-engineered winning solution of the xView2 competition, and gradually stripping away components. This way, we obtain a much simpler method, while retaining
Externí odkaz:
http://arxiv.org/abs/2401.17271
Autor:
Mohlin, David, Sullivan, Josephine
Estimating probability distributions which describe where an object is likely to be from camera data is a task with many applications. In this work we describe properties which we argue such methods should conform to. We also design a method which co
Externí odkaz:
http://arxiv.org/abs/2303.05245
Autor:
Gerard, Sebastian, Sullivan, Josephine
Domain-specific variants of contrastive learning can construct positive pairs from two distinct in-domain images, while traditional methods just augment the same image twice. For example, we can form a positive pair from two satellite images showing
Externí odkaz:
http://arxiv.org/abs/2211.13756
Autor:
Gamba, Matteo, Chmielewski-Anders, Adrian, Sullivan, Josephine, Azizpour, Hossein, Björkman, Mårten
The number of linear regions has been studied as a proxy of complexity for ReLU networks. However, the empirical success of network compression techniques like pruning and knowledge distillation, suggest that in the overparameterized setting, linear
Externí odkaz:
http://arxiv.org/abs/2202.11749
In this paper we describe a probabilistic method for estimating the position of an object along with its covariance matrix using neural networks. Our method is designed to be robust to outliers, have bounded gradients with respect to the network outp
Externí odkaz:
http://arxiv.org/abs/2111.10296
Autor:
Bujwid, Sebastian, Sullivan, Josephine
We study the impact of using rich and diverse textual descriptions of classes for zero-shot learning (ZSL) on ImageNet. We create a new dataset ImageNet-Wiki that matches each ImageNet class to its corresponding Wikipedia article. We show that merely
Externí odkaz:
http://arxiv.org/abs/2103.09669
Visual relationship detection is fundamental for holistic image understanding. However, the localization and classification of (subject, predicate, object) triplets remain challenging tasks, due to the combinatorial explosion of possible relationship
Externí odkaz:
http://arxiv.org/abs/2006.09562
Predicting attributes from face images in the wild is a challenging computer vision problem. To automatically describe face attributes from face containing images, traditionally one needs to cascade three technical blocks --- face localization, facia
Externí odkaz:
http://arxiv.org/abs/1602.03935
Predicting facial attributes from faces in the wild is very challenging due to pose and lighting variations in the real world. The key to this problem is to build proper feature representations to cope with these unfavourable conditions. Given the su
Externí odkaz:
http://arxiv.org/abs/1602.01827
This paper provides an extensive study on the availability of image representations based on convolutional networks (ConvNets) for the task of visual instance retrieval. Besides the choice of convolutional layers, we present an efficient pipeline exp
Externí odkaz:
http://arxiv.org/abs/1412.6574