1 code implementation • ECCV 2020 • Christian Simon, Piotr Koniusz, Richard Nock, Mehrtash Harandi
Inspired by optimization techniques, we propose a novel meta-learning algorithm with gradient modulation to encourage fast-adaptation of neural networks in the absence of abundant data.
no code implementations • 27 Mar 2024 • Yanshuo Wang, Ali Cheraghian, Zeeshan Hayder, Jie Hong, Sameera Ramasinghe, Shafin Rahman, David Ahmedt-Aristizabal, Xuesong Li, Lars Petersson, Mehrtash Harandi
Here, we propose a novel method that uses a backpropagation-free approach for TTA for the specific case of 3D data.
1 code implementation • 21 Mar 2024 • Minh-Tuan Tran, Trung Le, Xuan-May Le, Mehrtash Harandi, Dinh Phung
In this field, Data-Free Knowledge Transfer (DFKT) plays a crucial role in addressing catastrophic forgetting and data privacy problems.
1 code implementation • 21 Mar 2024 • Yihang Chen, Qianyi Wu, Jianfei Cai, Mehrtash Harandi, Weiyao Lin
3D Gaussian Splatting (3DGS) has emerged as a promising framework for novel view synthesis, boasting rapid rendering speed with high fidelity.
no code implementations • 11 Jan 2024 • Jing Wu, Trung Le, Munawar Hayat, Mehrtash Harandi
In this work, we introduce an unlearning algorithm for diffusion models.
no code implementations • 11 Jan 2024 • Jing Wu, Mehrtash Harandi
Machine unlearning has become a pivotal task to erase the influence of data from a trained model.
no code implementations • 18 Dec 2023 • Nilakshan Kunananthaseelan, Jing Zhang, Mehrtash Harandi
We introduce a language-grounded visual prompting method to adapt the visual encoder of vision-language models for downstream tasks.
no code implementations • 26 Oct 2023 • Yang Yi Poh, Ethan Grooby, Kenneth Tan, Lindsay Zhou, Arrabella King, Ashwin Ramanathan, Atul Malhotra, Mehrtash Harandi, Faezeh Marzbanrad
Inspired by the Conv-TasNet model, the proposed model has an encoder, decoder, and mask generator.
no code implementations • 5 Oct 2023 • Yanshuo Wang, Jie Hong, Ali Cheraghian, Shafin Rahman, David Ahmedt-Aristizabal, Lars Petersson, Mehrtash Harandi
DSS consists of dynamic thresholding, positive learning, and negative learning processes.
1 code implementation • 30 Sep 2023 • Minh-Tuan Tran, Trung Le, Xuan-May Le, Mehrtash Harandi, Quan Hung Tran, Dinh Phung
In this paper, we propose a novel Noisy Layer Generation method (NAYER) which relocates the random source from the input to a noisy layer and utilizes the meaningful constant label-text embedding (LTE) as the input.
no code implementations • 29 Sep 2023 • Tuan Truong, Hoang-Phi Nguyen, Tung Pham, Minh-Tuan Tran, Mehrtash Harandi, Dinh Phung, Trung Le
Motivated by this analysis, we introduce our algorithm, Riemannian Sharpness-Aware Minimization (RSAM).
no code implementations • ICCV 2023 • Jie Hong, Zeeshan Hayder, Junlin Han, Pengfei Fang, Mehrtash Harandi, Lars Petersson
Audio-visual zero-shot learning aims to classify samples consisting of a pair of corresponding audio and video sequences from classes that are not present during training.
Ranked #2 on GZSL Video Classification on ActivityNet-GZSL (cls)
no code implementations • 31 Jul 2023 • Kaushik Roy, Christian Simon, Peyman Moghadam, Mehrtash Harandi
To mitigate forgetting prior knowledge, we propose a novel knowledge distillation technique that takes into the account the manifold structure of the latent/output space of a neural network in learning novel tasks.
no code implementations • 31 Jul 2023 • Kaushik Roy, Peyman Moghadam, Mehrtash Harandi
To address the problem, we propose a distillation strategy named L3DMC that operates on mixed-curvature spaces to preserve the already-learned knowledge by modeling and maintaining complex geometrical structures.
1 code implementation • 21 Jul 2023 • Ruyi Zha, Xuelian Cheng, Hongdong Li, Mehrtash Harandi, ZongYuan Ge
We constrain the learned shape by tailoring multiple regularization strategies and disentangling geometry and appearance.
no code implementations • 1 Jun 2023 • Mevan Ekanayake, Zhifeng Chen, Mehrtash Harandi, Gary Egan, Zhaolin Chen
In Magnetic Resonance Imaging (MRI), image acquisitions are often undersampled in the measurement domain to accelerate the scanning process, at the expense of image quality.
no code implementations • 21 Apr 2023 • Pengfei Fang, Mehrtash Harandi, Trung Le, Dinh Phung
Hyperbolic geometry, a Riemannian manifold endowed with constant sectional negative curvature, has been considered an alternative embedding space in many learning scenarios, \eg, natural language processing, graph learning, \etc, as a result of its intriguing property of encoding the data's hierarchical structure (like irregular graph or tree-likeness data).
no code implementations • CVPR 2023 • Zhi Gao, Chen Xu, Feng Li, Yunde Jia, Mehrtash Harandi, Yuwei Wu
Our method dynamically expands the geometry of the underlying space to match growing geometric structures induced by new data, and prevents forgetting by keeping geometric structures of old data into account.
no code implementations • 12 Feb 2023 • Tung-Long Vuong, Trung Le, He Zhao, Chuanxia Zheng, Mehrtash Harandi, Jianfei Cai, Dinh Phung
Learning deep discrete latent presentations offers a promise of better symbolic and summarized abstractions that are more useful to subsequent downstream tasks.
no code implementations • ICCV 2023 • Samitha Herath, Basura Fernando, Ehsan Abbasnejad, Munawar Hayat, Shahram Khadivi, Mehrtash Harandi, Hamid Rezatofighi, Gholamreza Haffari
EBL can be used to improve the instance selection for a self-training task on the unlabelled target domain, and 2. alignment and normalizing energy scores can learn domain-invariant representations.
1 code implementation • 5 Dec 2022 • Jie Hong, Shi Qiu, Weihao Li, Saeed Anwar, Mehrtash Harandi, Nick Barnes, Lars Petersson
Specifically, we use the Unknown-Point Simulator to simulate out-of-distribution data in the training stage by manipulating the geometric context of partial known data.
no code implementations • 22 Nov 2022 • Yunyan Xing, Benjamin J. Meyer, Mehrtash Harandi, Tom Drummond, ZongYuan Ge
Medical imaging data, such as radiology images, are often multimorbidity; a single sample may have more than one pathology present.
1 code implementation • 19 Oct 2022 • Islam Nassar, Munawar Hayat, Ehsan Abbasnejad, Hamid Rezatofighi, Mehrtash Harandi, Gholamreza Haffari
We present LAVA, a simple yet effective method for multi-domain visual transfer learning with limited data.
1 code implementation • 9 Oct 2022 • Nicholas Rosa, Tom Drummond, Mehrtash Harandi
We demonstrate that our approach improves the fairness of AI models in varied task and dataset scenarios, whilst still maintaining a high level of classification accuracy.
1 code implementation • 16 Sep 2022 • Himashi Peiris, Munawar Hayat, Zhaolin Chen, Gary Egan, Mehrtash Harandi
As intensities of MRI volumes are inconsistent across institutes, it is essential to extract universal features of multi-modal MRIs to precisely segment brain tumors.
no code implementations • 14 Sep 2022 • Soumava Kumar Roy, Yan Han, Mehrtash Harandi, Lars Petersson
Deep Metric Learning algorithms aim to learn an efficient embedding space to preserve the similarity relationships among the input data.
1 code implementation • 13 Sep 2022 • Jing Wu, Munawar Hayat, Mingyi Zhou, Mehrtash Harandi
Federated Learning (FL) is a distributed learning paradigm that enhances users privacy by eliminating the need for clients to share raw, private data with the server.
no code implementations • 2 Aug 2022 • Jie Hong, Pengfei Fang, Weihao Li, Junlin Han, Lars Petersson, Mehrtash Harandi
Learning a latent embedding to understand the underlying nature of data distribution is often formulated in Euclidean spaces with zero curvature.
1 code implementation • 25 Jul 2022 • Xuelian Cheng, Yiran Zhong, Mehrtash Harandi, Tom Drummond, Zhiyong Wang, ZongYuan Ge
The self-attention mechanism, successfully employed with the transformer structure is shown promise in many computer vision tasks including image recognition, and object detection.
no code implementations • 18 Jul 2022 • Mevan Ekanayake, Kamlesh Pawar, Mehrtash Harandi, Gary Egan, Zhaolin Chen
Convolutional neural network (CNN) models are widely utilized for accelerated MRI reconstruction, but those models are limited in capturing global correlations due to the intrinsic locality of the convolution operation.
no code implementations • 15 Jun 2022 • Markus Hiller, Mehrtash Harandi, Tom Drummond
Inspired by the concept of preconditioning, we propose a novel method to increase adaptation speed for gradient-based meta-learning methods without incurring extra parameters.
1 code implementation • 15 Jun 2022 • Markus Hiller, Rongkai Ma, Mehrtash Harandi, Tom Drummond
Single image-level annotations only correctly describe an often small subset of an image's content, particularly when complex real-world scenes are depicted.
no code implementations • 23 Mar 2022 • Jie Hong, Weihao Li, Junlin Han, Jiyang Zheng, Pengfei Fang, Mehrtash Harandi, Lars Petersson
In this paper, we present and study a new image segmentation task, called Generalized Open-set Semantic Segmentation (GOSS).
1 code implementation • CVPR 2022 • Xuelian Cheng, Huan Xiong, Deng-Ping Fan, Yiran Zhong, Mehrtash Harandi, Tom Drummond, ZongYuan Ge
We propose a new video camouflaged object detection (VCOD) framework that can exploit both short-term dynamics and long-term temporal consistency to detect camouflaged objects from video frames.
Ranked #2 on Camouflaged Object Segmentation on Camouflaged Animal Dataset (using extra training data)
no code implementations • CVPR 2022 • Christian Simon, Masoud Faraki, Yi-Hsuan Tsai, Xiang Yu, Samuel Schulter, Yumin Suh, Mehrtash Harandi, Manmohan Chandraker
Humans have the ability to accumulate knowledge of new tasks in varying conditions, but deep neural networks often suffer from catastrophic forgetting of previously learned knowledge after learning a new task.
1 code implementation • 11 Jan 2022 • Himashi Peiris, Zhaolin Chen, Gary Egan, Mehrtash Harandi
In this concept, the 3D segmentation network learns from dual reciprocal adversarial learning approaches.
1 code implementation • 7 Dec 2021 • Rongkai Ma, Pengfei Fang, Gil Avraham, Yan Zuo, Tianyu Zhu, Tom Drummond, Mehrtash Harandi
A principle way of achieving few-shot learning is to realize a model that can rapidly adapt to the context of a given task.
no code implementations • 3 Dec 2021 • Rongkai Ma, Pengfei Fang, Tom Drummond, Mehrtash Harandi
To this end, we formulate the metric as a weighted sum on the tangent bundle of the hyperbolic space and develop a mechanism to obtain the weights adaptively and based on the constellation of the points.
1 code implementation • 26 Nov 2021 • Himashi Peiris, Munawar Hayat, Zhaolin Chen, Gary Egan, Mehrtash Harandi
We propose a Transformer architecture for volumetric segmentation, a challenging task that requires keeping a complex balance in encoding local and global spatial cues, and preserving information along all axes of the volume.
no code implementations • 22 Nov 2021 • Jing Zhang, Yuchao Dai, Mehrtash Harandi, Yiran Zhong, Nick Barnes, Richard Hartley
Uncertainty estimation has been extensively studied in recent literature, which can usually be classified as aleatoric uncertainty and epistemic uncertainty.
no code implementations • 12 Nov 2021 • Tianyu Zhu, Rongkai Ma, Mehrtash Harandi, Tom Drummond
A segmentation model cannot easily learn from prior information given in the visual tracking scenario.
no code implementations • 26 Oct 2021 • Christian Simon, Piotr Koniusz, Mehrtash Harandi
Even with the luxury of having abundant data, multi-label classification is widely known to be a challenging task to address.
no code implementations • 23 Oct 2021 • Christian Simon, Piotr Koniusz, Lars Petersson, Yan Han, Mehrtash Harandi
Our empirical evaluations show that the noise injecting operation does not degrade the performance of the NAS algorithm if the data is indeed clean.
1 code implementation • 13 Oct 2021 • Jing Zhang, Yuchao Dai, Mochu Xiang, Deng-Ping Fan, Peyman Moghadam, Mingyi He, Christian Walder, Kaihao Zhang, Mehrtash Harandi, Nick Barnes
Deep neural networks can be roughly divided into deterministic neural networks and stochastic neural networks. The former is usually trained to achieve a mapping from input space to output space via maximum likelihood estimation for the weights, which leads to deterministic predictions during testing.
no code implementations • 29 Sep 2021 • Xuelian Cheng, Huan Xiong, Deng-Ping Fan, Yiran Zhong, Mehrtash Harandi, Tom Drummond, ZongYuan Ge
The proposed SLT-Net leverages on both short-term dynamics and long-term temporal consistency to detect concealed objects in continuous video frames.
no code implementations • 29 Sep 2021 • Nicholas I-Hsien Kuo, Mehrtash Harandi, Nicolas Fourrier, Gabriela Ferraro, Christian Walder, Hanna Suominen
Neural networks usually excel in learning a single task.
no code implementations • 20 Sep 2021 • Jieming Zhou, Tong Zhang, Pengfei Fang, Lars Petersson, Mehrtash Harandi
The core concept of GNNs is to find a representation by recursively aggregating the representations of a central node and those of its neighbors.
1 code implementation • 25 Aug 2021 • Himashi Peiris, Zhaolin Chen, Gary Egan, Mehrtash Harandi
Segmentation of images is a long-standing challenge in medical AI.
no code implementations • 10 May 2021 • Ujjal Kr Dutta, Mehrtash Harandi, Chellu Chandra Sekhar
In this paper, we address the problem of Semi-Supervised DML (SSDML) that tries to learn a metric using a few labeled examples, and abundantly available unlabeled examples.
1 code implementation • 22 Apr 2021 • Jing Wu, Mingyi Zhou, Ce Zhu, Yipeng Liu, Mehrtash Harandi, Li Li
Recently, adversarial attack methods have been developed to challenge the robustness of machine learning models.
1 code implementation • CVPR 2021 • Christian Simon, Piotr Koniusz, Mehrtash Harandi
This in practice is done by minimizing the dissimilarity between current and previous responses of the network one way or another.
no code implementations • 13 Apr 2021 • Anoop Cherian, Panagiotis Stanitsas, Jue Wang, Mehrtash Harandi, Vassilios Morellas, Nikolaos Papanikolopoulos
There exist several similarity measures for comparing SPD matrices with documented benefits.
no code implementations • CVPR 2021 • Jie Hong, Pengfei Fang, Weihao Li, Tong Zhang, Christian Simon, Mehrtash Harandi, Lars Petersson
Few-shot learning aims to correctly recognize query samples from unseen classes given a limited number of support samples, often by relying on global embeddings of images.
no code implementations • CVPR 2021 • Ali Cheraghian, Shafin Rahman, Pengfei Fang, Soumava Kumar Roy, Lars Petersson, Mehrtash Harandi
Few-shot class incremental learning (FSCIL) portrays the problem of learning new concepts gradually, where only a few examples per concept are available to the learner.
1 code implementation • 6 Mar 2021 • Nicholas I-Hsien Kuo, Mehrtash Harandi, Nicolas Fourrier, Christian Walder, Gabriela Ferraro, Hanna Suominen
Neural networks suffer from catastrophic forgetting and are unable to sequentially learn new tasks without guaranteed stationarity in data distribution.
no code implementations • 28 Feb 2021 • Lie Ju, Xin Wang, Lin Wang, Dwarikanath Mahapatra, Xin Zhao, Mehrtash Harandi, Tom Drummond, Tongliang Liu, ZongYuan Ge
In this paper, we systematically discuss and define the two common types of label noise in medical images - disagreement label noise from inconsistency expert opinions and single-target label noise from wrong diagnosis record.
no code implementations • ICCV 2021 • Pengfei Fang, Mehrtash Harandi, Lars Petersson
However, working in hyperbolic spaces is not without difficulties as a result of its curved geometry (e. g., computing the Frechet mean of a set of points requires an iterative algorithm).
no code implementations • ICCV 2021 • Zhi Gao, Yuwei Wu, Yunde Jia, Mehrtash Harandi
Few-shot learning describes the challenging problem of recognizing samples from unseen classes given very few labeled examples.
no code implementations • 1 Jan 2021 • Nicholas I-Hsien Kuo, Mehrtash Harandi, Nicolas Fourrier, Christian Walder, Gabriela Ferraro, Hanna Suominen
Catastrophic forgetting occurs when a neural network is trained sequentially on multiple tasks – its weights will be continuously modified and as a result, the network will lose its ability in solving a previous task.
no code implementations • ICCV 2021 • Ali Cheraghian, Shafin Rahman, Sameera Ramasinghe, Pengfei Fang, Christian Simon, Lars Petersson, Mehrtash Harandi
In this paper, we propose addressing this problem using a mixture of subspaces.
no code implementations • 10 Dec 2020 • Jing Zhang, Yuchao Dai, Xin Yu, Mehrtash Harandi, Nick Barnes, Richard Hartley
Existing deep neural network based salient object detection (SOD) methods mainly focus on pursuing high network accuracy.
no code implementations • 2 Nov 2020 • Pengfei Fang, Pan Ji, Lars Petersson, Mehrtash Harandi
Modern video person re-identification (re-ID) machines are often trained using a metric learning approach, supervised by a triplet loss.
1 code implementation • NeurIPS 2020 • Xuelian Cheng, Yiran Zhong, Mehrtash Harandi, Yuchao Dai, Xiaojun Chang, Tom Drummond, Hongdong Li, ZongYuan Ge
To reduce the human efforts in neural network design, Neural Architecture Search (NAS) has been applied with remarkable success to various high-level vision tasks such as classification and semantic segmentation.
Ranked #2 on Stereo Disparity Estimation on Scene Flow
no code implementations • 7 Oct 2020 • Pengfei Fang, Pan Ji, Jieming Zhou, Lars Petersson, Mehrtash Harandi
Full attention, which generates an attention value per element of the input feature maps, has been successfully demonstrated to be beneficial in visual tasks.
no code implementations • 22 Aug 2020 • Ujjal Kr Dutta, Mehrtash Harandi, Chellu Chandra Sekhar
As obtaining class labels in all applications is not feasible, we propose an unsupervised approach that learns a metric without making use of class labels.
1 code implementation • 18 Jul 2020 • Nicholas I-Hsien Kuo, Mehrtash Harandi, Nicolas Fourrier, Christian Walder, Gabriela Ferraro, Hanna Suominen
Learning to learn (L2L) trains a meta-learner to assist the learning of a task-specific base learner.
no code implementations • 17 Jun 2020 • Jieming Zhou, Soumava Kumar Roy, Pengfei Fang, Mehrtash Harandi, Lars Petersson
Deep neural networks need to make robust inference in the presence of occlusion, background clutter, pose and viewpoint variations -- to name a few -- when the task of person re-identification is considered.
no code implementations • 23 Mar 2020 • Lie Ju, Xin Wang, Quan Zhou, Hu Zhu, Mehrtash Harandi, Paul Bonnington, Tom Drummond, ZongYuan Ge
We design a regularisation technique to regulate the domain adaptation.
no code implementations • 27 Feb 2020 • Ujjal Kr Dutta, Mehrtash Harandi, Chellu Chandra Sekhar
In this paper, we revamp the forgotten classical Semi-Supervised Distance Metric Learning (SSDML) problem from a Riemannian geometric lens, to leverage stochastic optimization within a end-to-end deep framework.
no code implementations • 17 Dec 2019 • Ujjal Kr Dutta, Mehrtash Harandi, Chandra Sekhar Chellu
This restricts their applicability for large datasets in new applications where obtaining labels require extensive manual efforts and domain knowledge.
no code implementations • ICLR 2019 • Christian Simon, Piotr Koniusz, Mehrtash Harandi
Generalization from limited examples, usually studied under the umbrella of meta-learning, equips learning techniques with the ability to adapt quickly in dynamical environments and proves to be an essential aspect of lifelong learning.
no code implementations • 24 Apr 2019 • Tong Zhang, Pan Ji, Mehrtash Harandi, Wenbing Huang, Hongdong Li
We introduce the Neural Collaborative Subspace Clustering, a neural model that discovers clusters of data points drawn from a union of low-dimensional subspaces.
no code implementations • 2 Nov 2018 • Tong Zhang, Pan Ji, Mehrtash Harandi, Richard Hartley, Ian Reid
In this paper, we introduce a method that simultaneously learns an embedding space along subspaces within it to minimize a notion of reconstruction error, thus addressing the problem of subspace clustering in an end-to-end learning paradigm.
no code implementations • ECCV 2018 • Piotr Koniusz, Yusuf Tas, Hongguang Zhang, Mehrtash Harandi, Fatih Porikli, Rui Zhang
To achieve robust baselines, we build on a recent approach that aligns per-class scatter matrices of the source and target CNN streams.
no code implementations • CVPR 2018 • Soumava Kumar Roy, Zakaria Mhammedi, Mehrtash Harandi
In this paper, we extend some popular optimization algorithm to the Riemannian (constrained) setting.
no code implementations • 16 Apr 2018 • Yao Lu, Mehrtash Harandi, Richard Hartley, Razvan Pascanu
Advanced optimization algorithms such as Newton method and AdaGrad benefit from second order derivative or second order statistics to achieve better descent directions and faster convergence rates.
no code implementations • CVPR 2018 • Jing Zhang, Tong Zhang, Yuchao Dai, Mehrtash Harandi, Richard Hartley
Such supervision, while labor-intensive and not always possible, tends to hinder the generalization ability of the learned models.
no code implementations • 20 Feb 2018 • Yao Lu, Jack Valmadre, Heng Wang, Juho Kannala, Mehrtash Harandi, Philip H. S. Torr
State-of-the-art neural network models estimate large displacement optical flow in multi-resolution and use warping to propagate the estimation between two resolutions.
no code implementations • 4 Feb 2018 • Piotr Koniusz, Yusuf Tas, Hongguang Zhang, Mehrtash Harandi, Fatih Porikli, Rui Zhang
To achieve robust baselines, we build on a recent approach that aligns per-class scatter matrices of the source and target CNN streams [15].
no code implementations • NeurIPS 2017 • Wenbing Huang, Mehrtash Harandi, Tong Zhang, Lijie Fan, Fuchun Sun, Junzhou Huang
Linear Dynamical Systems (LDSs) are fundamental tools for modeling spatio-temporal data in various disciplines.
no code implementations • ICCV 2017 • Anoop Cherian, Panagiotis Stanitsas, Mehrtash Harandi, Vassilios Morellas, Nikolaos Papanikolopoulos
Symmetric positive definite (SPD) matrices are useful for capturing second-order statistics of visual data.
no code implementations • 5 Aug 2017 • Anoop Cherian, Panagiotis Stanitsas, Mehrtash Harandi, Vassilios Morellas, Nikolaos Papanikolopoulos
Symmetric positive definite (SPD) matrices are useful for capturing second-order statistics of visual data.
no code implementations • ICML 2017 • Mehrtash Harandi, Mathieu Salzmann, Richard Hartley
To be tractable and robust to data noise, existing metric learning algorithms commonly rely on PCA as a pre-processing step.
no code implementations • CVPR 2017 • Anoop Cherian, Basura Fernando, Mehrtash Harandi, Stephen Gould
Most popular deep models for action recognition split video sequences into short sub-sequences consisting of a few frames; frame-based features are then pooled for recognizing the activity.
no code implementations • CVPR 2017 • Samitha Herath, Mehrtash Harandi, Fatih Porikli
This paper introduces a learning scheme to construct a Hilbert space (i. e., a vector space along its inner product) to address both unsupervised and semi-supervised domain adaptation problems.
no code implementations • 17 Nov 2016 • Mehrtash Harandi, Basura Fernando
This paper introduces an extension of the backpropagation algorithm that enables us to have layers with constrained weights in a deep network.
Dimensionality Reduction Fine-Grained Image Classification +1
no code implementations • 3 Aug 2016 • Wenbing Huang, Fuchun Sun, Lele Cao, Mehrtash Harandi
We then devise efficient algorithms to perform sparse coding and dictionary learning on the space of infinite-dimensional subspaces.
no code implementations • CVPR 2016 • Wenbing Huang, Fuchun Sun, Lele Cao, Deli Zhao, Huaping Liu, Mehrtash Harandi
To enhance the performance of LDSs, in this paper, we address the challenging issue of performing sparse coding on the space of LDSs, where both data and dictionary atoms are LDSs.
no code implementations • 20 May 2016 • Mehrtash Harandi, Mathieu Salzmann, Richard Hartley
This lets us formulate dimensionality reduction as the problem of finding a projection that yields a low-dimensional manifold either with maximum discriminative power in the supervised scenario, or with maximum variance of the data in the unsupervised one.
no code implementations • 16 May 2016 • Samitha Herath, Mehrtash Harandi, Fatih Porikli
Understanding human actions in visual data is tied to advances in complementary research areas including object recognition, human dynamics, domain adaptation and semantic segmentation.
no code implementations • ICCV 2015 • Mehrtash Harandi, Mathieu Salzmann, Mahsa Baktashmotlagh
State-of-the-art image-set matching techniques typically implicitly model each image-set with a Gaussian distribution.
no code implementations • CVPR 2016 • Mehrtash Harandi, Mathieu Salzmann, Fatih Porikli
Vectors of Locally Aggregated Descriptors (VLAD) have emerged as powerful image/video representations that compete with or even outperform state-of-the-art approaches on many challenging visual recognition tasks.
no code implementations • CVPR 2015 • Mehrtash Harandi, Mathieu Salzmann
While sparse coding on non-flat Riemannian manifolds has recently become increasingly popular, existing solutions either are dedicated to specific manifolds, or rely on optimization problems that are difficult to solve, especially when it comes to dictionary learning.
no code implementations • CVPR 2013 • Sadeep Jayasumana, Richard Hartley, Mathieu Salzmann, Hongdong Li, Mehrtash Harandi
To encode the geometry of the manifold in the mapping, we introduce a family of provably positive definite kernels on the Riemannian manifold of SPD matrices.
no code implementations • CVPR 2014 • Sadeep Jayasumana, Richard Hartley, Mathieu Salzmann, Hongdong Li, Mehrtash Harandi
We tackle the problem of optimizing over all possible positive definite radial kernels on Riemannian manifolds for classification.
no code implementations • 13 Dec 2014 • Sadeep Jayasumana, Mathieu Salzmann, Hongdong Li, Mehrtash Harandi
We propose a framework for 2D shape analysis using positive definite kernels defined on Kendall's shape manifold.
no code implementations • 30 Nov 2014 • Sadeep Jayasumana, Richard Hartley, Mathieu Salzmann, Hongdong Li, Mehrtash Harandi
We then use the proposed framework to identify positive definite kernels on two specific manifolds commonly encountered in computer vision: the Riemannian manifold of symmetric positive definite matrices and the Grassmann manifold, i. e., the Riemannian manifold of linear subspaces of a Euclidean space.
no code implementations • 30 Aug 2014 • Mehrtash Harandi, Richard Hartley, Brian Lovell, Conrad Sanderson
This paper introduces sparse coding and dictionary learning for Symmetric Positive Definite (SPD) matrices, which are often used in machine learning, computer vision and related areas.
no code implementations • 30 Aug 2014 • Mehrtash Harandi, Mathieu Salzmann
In contrast, here, we study the problem of performing coding in a high-dimensional Hilbert space, where the classes are expected to be more easily separable.
no code implementations • CVPR 2014 • Mehrtash Harandi, Mathieu Salzmann, Fatih Porikli
We introduce an approach to computing and comparing Covariance Descriptors (CovDs) in infinite-dimensional spaces.
no code implementations • 4 Mar 2014 • Azadeh Alavi, Yan Yang, Mehrtash Harandi, Conrad Sanderson
The use of similarity vectors is in contrast to the traditional approach of embedding manifolds into tangent spaces, which can suffer from representing the manifold structure inaccurately.
no code implementations • 31 Jan 2014 • Mehrtash Harandi, Richard Hartley, Chunhua Shen, Brian Lovell, Conrad Sanderson
With the aim of building a bridge between the two realms, we address the problem of sparse coding and dictionary learning over the space of linear subspaces, which form Riemannian structures known as Grassmann manifolds.
no code implementations • 18 Oct 2013 • Mehrtash Harandi, Conrad Sanderson, Chunhua Shen, Brian C. Lovell
Recent advances in computer vision and machine learning suggest that a wide range of problems can be addressed more appropriately by considering non-Euclidean geometry.