no code implementations • 22 Mar 2023 • Frances Fengyi Yang, Michele Sasdelli, Tat-Jun Chin
This leads to a strategy to train MLPs with quantum annealers as a sampling engine.
1 code implementation • CVPR 2022 • Anh-Dzung Doan, Michele Sasdelli, David Suter, Tat-Jun Chin
While our usage of quantum computing does not surmount the fundamental intractability of robust fitting, by providing error bounds our algorithm is a practical improvement over randomised heuristics.
no code implementations • 25 Jan 2022 • Gerard Snaauw, Michele Sasdelli, Gabriel Maicas, Stephan Lau, Johan Verjans, Mark Jenkinson, Gustavo Carneiro
We propose guiding the training of a deep learning-based registration method with MI estimation between an image-pair in an end-to-end trainable network.
no code implementations • 3 Dec 2021 • Andrew Du, Yee Wei Law, Michele Sasdelli, Bo Chen, Ken Clarke, Michael Brown, Tat-Jun Chin
In fact, advanced EO satellites perform deep learning-based cloud detection on board the satellites and downlink only clear-sky data to save precious bandwidth.
1 code implementation • 26 Aug 2021 • Andrew Du, Bo Chen, Tat-Jun Chin, Yee Wei Law, Michele Sasdelli, Ramesh Rajasegaran, Dillon Campbell
In this work, we demonstrate one of the first efforts on physical adversarial attacks on aerial imagery, whereby adversarial patches were optimised, fabricated and installed on or near target objects (cars) to significantly reduce the efficacy of an object detector applied on overhead images.
no code implementations • 5 Jul 2021 • Michele Sasdelli, Tat-Jun Chin
Quantum annealing is a promising paradigm for building practical quantum computers.
no code implementations • 1 Jan 2021 • Michele Sasdelli, Thalaiyasingam Ajanthan, Tat-Jun Chin, Gustavo Carneiro
Then, we empirically show that for a large range of learning rates, SGD traverses the loss landscape across regions with largest eigenvalue of the Hessian similar to the inverse of the learning rate.
no code implementations • 14 Aug 2019 • Rafael Felix, Ben Harwood, Michele Sasdelli, Gustavo Carneiro
Most state-of-the-art GZSL approaches rely on a mapping between latent visual and semantic spaces without considering if a particular sample belongs to the set of seen or unseen classes.
no code implementations • 6 Aug 2019 • Rafael Felix, Ben Harwood, Michele Sasdelli, Gustavo Carneiro
In the context of GZSL, semantic information is non-visual data such as a text description of both seen and unseen classes.
no code implementations • 12 Feb 2019 • Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli
When the confidence is low, we avoid updating the object's position through the feature map; instead, the tracker passes to a single-frame failure mode, during which the patch's low-level visual content is used to swiftly update the object's position, before recovering from the target loss in the next frame.
no code implementations • 15 Jan 2019 • Rafael Felix, Michele Sasdelli, Ian Reid, Gustavo Carneiro
In this paper, we mitigate these issues by proposing a new GZSL method based on multi-modal training and testing processes, where the optimization explicitly promotes a balanced classification accuracy between seen and unseen classes.
no code implementations • 7 Nov 2018 • Roshanak Zakizadeh, Yu Qian, Michele Sasdelli, Eduard Vazquez
In this paper, we present a method for instance ranking and retrieval at fine-grained level based on the global features extracted from a multi-attribute recognition model which is not dependent on landmarks information or part-based annotations.
1 code implementation • 31 Jul 2018 • Roshanak Zakizadeh, Michele Sasdelli, Yu Qian, Eduard Vazquez
After selecting categories with sufficient number of images for training, we remove very scarce attributes and merge the duplicate ones in each category, then we clean the dataset based on the new list of attributes.
no code implementations • 20 Jun 2018 • Alessandro Bay, Panagiotis Sidiropoulos, Eduard Vazquez, Michele Sasdelli
In the case of a low confidence, the position update is rejected and the tracker passes to a single-frame failure mode, during which the patch low-level visual content is used to swiftly update the object position, before recovering from the target loss in the next frame.
no code implementations • 19 Jun 2018 • Roshanak Zakizadeh, Michele Sasdelli, Yu Qian, Eduard Vazquez
In this paper, we address the extraction of the fine-grained attributes of an instance as a `multi-attribute classification' problem.