31 Hits in 0.93 sec

Geometric Disentanglement for Generative Latent Shape Models [article]

Tristan Aumentado-Armstrong, Stavros Tsogkas, Allan Jepson, Sven Dickinson
2019 arXiv   pre-print
Sven Dickinson, Allan Jepson, and Stavros Tsogkas contributed in their capacity as Professors and Postdoc at the University of Toronto, respectively.  ... 
arXiv:1908.06386v1 fatcat:dtjbjhjxz5dlnjkth5f3o4tkaa

DeepFlux for Skeletons in the Wild [article]

Yukang Wang, Yongchao Xu, Stavros Tsogkas, Xiang Bai, Sven Dickinson, Kaleem Siddiqi
2018 arXiv   pre-print
Tsogkas and Kokkinos [45] extract hand-designed features at each pixel and train a classifier for symmetry detection.  ... 
arXiv:1811.12608v1 fatcat:pahjsjqtfbhzbmaawn4yd3ktxq

AMAT: Medial Axis Transform for Natural Images [article]

Stavros Tsogkas, Sven Dickinson
2017 arXiv   pre-print
Our code and annotations are available at .  ...  Tsogkas and Kokkinos use multiple instance learning (MIL) to deal with the unknown scale and orientation during training [49] , while Shen et al. adapt a CNN with side outputs [53] for object skeleton  ... 
arXiv:1703.08628v2 fatcat:lnzslfhxsbbupgq3oekshmxwai

Disentangling Geometric Deformation Spaces in Generative Latent Shape Models [article]

Tristan Aumentado-Armstrong, Stavros Tsogkas, Sven Dickinson, Allan Jepson
2021 arXiv   pre-print
1 University of Toronto 2 Samsung Toronto AI Research Center 3 Vector Institute for Artificial Intelligence Disclaimer: Tristan Aumentado-Armstrong and Stavros Tsogkas contributed to this article in their  ... 
arXiv:2103.00142v1 fatcat:oefxuozl5bewtcnbvqoe66ccze

GIFT: Generalizable Interaction-aware Functional Tool Affordances without Labels [article]

Dylan Turpin, Liquan Wang, Stavros Tsogkas, Sven Dickinson, Animesh Garg
2021 arXiv   pre-print
Tool use requires reasoning about the fit between an object's affordances and the demands of a task. Visual affordance learning can benefit from goal-directed interaction experience, but current techniques rely on human labels or expert demonstrations to generate this data. In this paper, we describe a method that grounds affordances in physical interactions instead, thus removing the need for human labels or expert policies. We use an efficient sampling-based method to generate successful
more » ... ctories that provide contact data, which are then used to reveal affordance representations. Our framework, GIFT, operates in two phases: first, we discover visual affordances from goal-directed interaction with a set of procedurally generated tools; second, we train a model to predict new instances of the discovered affordances on novel tools in a self-supervised fashion. In our experiments, we show that GIFT can leverage a sparse keypoint representation to predict grasp and interaction points to accommodate multiple tasks, such as hooking, reaching, and hammering. GIFT outperforms baselines on all tasks and matches a human oracle on two of three tasks using novel tools.
arXiv:2106.14973v1 fatcat:g4jzc6pnzrcclmfzwpimtuja5a

Learning-Based Symmetry Detection in Natural Images [chapter]

Stavros Tsogkas, Iasonas Kokkinos
2012 Lecture Notes in Computer Science  
In this work we propose a learning-based approach to symmetry detection in natural images. We focus on ribbon-like structures, i.e. contours marking local and approximate reflection symmetry and make three contributions to improve their detection. First, we create and make publicly available a ground-truth dataset for this task by building on the Berkeley Segmentation Dataset. Second, we extract features representing multiple complementary cues, such as grayscale structure, color, texture, and
more » ... pectral clustering information. Third, we use supervised learning to learn how to combine these cues, and employ MIL to accommodate the unknown scale and orientation of the symmetric structures. We systematically evaluate the performance contribution of each individual component in our pipeline, and demonstrate that overall we consistently improve upon results obtained using existing alternatives.
doi:10.1007/978-3-642-33786-4_4 fatcat:ts3ommyzq5dfpjxc6mrprjtffy

Cycle-Consistent Generative Rendering for 2D-3D Modality Translation [article]

Tristan Aumentado-Armstrong, Alex Levinshtein, Stavros Tsogkas, Konstantinos G. Derpanis, Allan D. Jepson
2020 arXiv   pre-print
For humans, visual understanding is inherently generative: given a 3D shape, we can postulate how it would look in the world; given a 2D image, we can infer the 3D structure that likely gave rise to it. We can thus translate between the 2D visual and 3D structural modalities of a given object. In the context of computer vision, this corresponds to a learnable module that serves two purposes: (i) generate a realistic rendering of a 3D object (shape-to-image translation) and (ii) infer a
more » ... 3D shape from an image (image-to-shape translation). In this paper, we learn such a module while being conscious of the difficulties in obtaining large paired 2D-3D datasets. By leveraging generative domain translation methods, we are able to define a learning algorithm that requires only weak supervision, with unpaired data. The resulting model is not only able to perform 3D shape, pose, and texture inference from 2D images, but can also generate novel textured 3D shapes and renders, similar to a graphics pipeline. More specifically, our method (i) infers an explicit 3D mesh representation, (ii) utilizes example shapes to regularize inference, (iii) requires only an image mask (no keypoints or camera extrinsics), and (iv) has generative capabilities. While prior work explores subsets of these properties, their combination is novel. We demonstrate the utility of our learned representation, as well as its performance on image generation and unpaired 3D shape inference tasks.
arXiv:2011.08026v1 fatcat:irpf4btl6vdvvldr6zgqgfwxpi

Sub-cortical brain structure segmentation using F-CNN's [article]

Mahsa Shakeri, Stavros Tsogkas, Enzo Ferrante (CVN, GALEN), Sarah Lippe, Samuel Kadoury, Nikos Paragios, Iasonas Kokkinos
2016 arXiv   pre-print
Code, computed probability maps, and more results can be found at brainseg.  ... 
arXiv:1602.02130v1 fatcat:xsc2nhefmzfzbn5jgmul3km4wa

Appearance Shock Grammar for Fast Medial Axis Extraction from Real Images [article]

Charles-Olivier Dufresne Camaro, Morteza Rezanejad, Stavros Tsogkas, Kaleem Siddiqi, Sven Dickinson
2020 arXiv   pre-print
A completely different, unsupervised approach, the AMAT, was proposed by Tsogkas and Dickinson [36] .  ...  Departing from this trend, Tsogkas and Dickinson defined the first complete MAT for color images, formulating medial axis extraction as a set cover problem [36] .  ... 
arXiv:2004.02677v1 fatcat:vohghovsererffnewmmzklfkqa

Prior-Based Coregistration and Cosegmentation [chapter]

Mahsa Shakeri, Enzo Ferrante, Stavros Tsogkas, Sarah Lippé, Samuel Kadoury, Iasonas Kokkinos, Nikos Paragios
2016 Lecture Notes in Computer Science  
We propose a modular and scalable framework for dense coregistration and cosegmentation with two key characteristics: first, we substitute ground truth data with the semantic map output of a classifier; second, we combine this output with population deformable registration to improve both alignment and segmentation. Our approach deforms all volumes towards consensus, taking into account image similarities and label consistency. Our pipeline can incorporate any classifier and similarity metric.
more » ... esults on two datasets, containing annotations of challenging brain structures, demonstrate the potential of our method.
doi:10.1007/978-3-319-46723-8_61 fatcat:ommhkeaafjhe7opwagh2fsay3u

Segmentation-Aware Deformable Part Models

Eduard Trulls, Stavros Tsogkas, Iasonas Kokkinos, Alberto Sanfeliu, Francesc Moreno-Noguer
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
In this work we propose a technique to combine bottomup segmentation, coming in the form of SLIC superpixels, with sliding window detectors, such as Deformable Part Models (DPMs). The merit of our approach lies in 'cleaning up' the lowlevel HOG features by exploiting the spatial support of SLIC superpixels; this can be understood as using segmentation to split the feature variation into object-specific and background changes. Rather than committing to a single segmentation we use a large pool
more » ... SLIC superpixels and combine them in a scale-, position-and object-dependent manner to build soft segmentation masks. The segmentation masks can be computed fast enough to repeat this process over every candidate window, during training and detection, for both the root and part filters of DPMs. We use these masks to construct enhanced, backgroundinvariant features to train DPMs. We test our approach on the PASCAL VOC 2007, outperforming the standard DPM in 17 out of 20 classes, yielding an average increase of 1.7% AP. Additionally, we demonstrate the robustness of this approach, extending it to dense SIFT descriptors for large displacement optical flow.
doi:10.1109/cvpr.2014.29 dblp:conf/cvpr/TrullsTKSM14 fatcat:gefw6vemdfddjoqxeezu65mvra

Learning Compositional Shape Priors for Few-Shot 3D Reconstruction [article]

Mateusz Michalkiewicz, Stavros Tsogkas, Sarah Parisot, Mahsa Baktashmotlagh, Anders Eriksson, Eugene Belilovsky
2021 arXiv   pre-print
. * Stavros Tsogkas and Sarah Parisot contributed to this article in their personal capacity as an Adjunct Professor at the University of Toronto and Visiting Scholar at Mila, respectively.  ...  Tsogkas is with the University of Toronto and Samsung AI Research Center, Toronto. • E.  ... 
arXiv:2106.06440v2 fatcat:ncnobbj7mndbpdgmuzz6cefumq

Accurate Human-Limb Segmentation in RGB-D Images for Intelligent Mobility Assistance Robots

Siddhartha Chandra, Stavros Tsogkas, Iasonas Kokkinos
2015 2015 IEEE International Conference on Computer Vision Workshop (ICCVW)  
Mobility impairment is one of the biggest challenges faced by elderly people in today's society. The inability to move about freely poses severe restrictions on their independence and general quality of life. This work is dedicated to developing intelligent robotic platforms that assist users to move without requiring a human attendant. This work was done in the context of an EU project involved in developing an intelligent robot for elderly user assistance. The robot is equipped with a Kinect
more » ... ensor, and the visioncomponent of the project has the responsibility of locating the user, estimating the user's pose, and recognizing gestures by the user. All these goals can take advantage of a method that accurately segments human-limbs in the colour (RGB) and depth (D) images captured by the Kinect sensor. We exploit recent advances in deep-learning to develop a system that performs accurate semantic segmentation of human limbs using colour and depth images. Our novel technical contributions are the following: 1) we describe a scheme for manual annotation of videos, that eliminates the need to annotate segmentation masks in every single frame; 2) we extend a state of the art deep learning system for semantic segmentation, to exploit diverse RGB and depth data, in a single framework for training and testing; 3) we evaluate different variants of our system and demonstrate promising performance, as well the contribution of diverse data, on our in-house Human-Limb dataset. Our method is very efficient, running at 8 frames per second on a GPU.
doi:10.1109/iccvw.2015.64 dblp:conf/iccvw/ChandraTK15 fatcat:5x63ha4nzbci5jw5diteyrslh4

2017 ICCV Challenge: Detecting Symmetry in the Wild

Christopher Funk, Seungkyu Lee, Martin R. Oswald, Stavros Tsogkas, Wei Shen, Andrea Cohen, Sven Dickinson, Yanxi Liu
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
The bottom row image is from the BMAX500 dataset and the algorithm is from Tsogkas and Kokkinos [47] . symmetry or medial axis basis.  ...  We also compare against Loy and Ek- lundh's algorithm [26] (LE), Tsogkas and Kokkinos's algorithm [47] (MIL), the Structured Random Forest method of Teo et al.  ... 
doi:10.1109/iccvw.2017.198 dblp:conf/iccvw/FunkLOTSCDL17 fatcat:jbj2fad3rrbcrl2spcdzn2gwl4

Sub-cortical brain structure segmentation using F-CNN'S

Mahsa Shakeri, Stavros Tsogkas, Enzo Ferrante, Sarah Lippe, Samuel Kadoury, Nikos Paragios, Iasonas Kokkinos
2016 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI)  
Code, computed probability maps, and more results can be found at brainseg.  ... 
doi:10.1109/isbi.2016.7493261 dblp:conf/isbi/ShakenTFLKPK16 fatcat:pqgyiy6skbfunfrd34vuzqz5ly
« Previous Showing results 1 — 15 out of 31 results