Filters








68,294 Hits in 4.0 sec

NucMM Dataset: 3D Neuronal Nuclei Instance Segmentation at Sub-Cubic Millimeter Scale [article]

Zudi Lin, Donglai Wei, Mariela D. Petkova, Yuelong Wu, Zergham Ahmed, Krishna Swaroop K, Silin Zou, Nils Wendt, Jonathan Boulanger-Weill, Xueying Wang, Nagaraju Dhanyasi, Ignacio Arganda-Carreras (+3 others)
2021 arXiv   pre-print
To tackle the challenges, we propose a novel hybrid-representation learning model that combines the merits of foreground mask, contour map, and signed distance transform to produce high-quality 3D masks  ...  Segmenting 3D cell nuclei from microscopy image volumes is critical for biological and clinical analysis, enabling the study of cellular expression patterns and cell lineages.  ...  We thank Daniel Franco-Barranco for setting up the challenge using NucMM.  ... 
arXiv:2107.05840v1 fatcat:zrcanxxsxrcwdhzssizyts3bcq

Learning sign language by watching TV (using weakly aligned subtitles)

P. Buehler, A. Zisserman, M. Everingham
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
; (ii) we show that by optimizing a scoring function based on multiple instance learning, we are able to extract the sign of interest from hours of signing footage, despite the very weak and noisy supervision  ...  The contributions are: (i) we propose a distance function to match signing sequences which includes the trajectory of both hands, the hand shape and orientation, and properly models the case of hands touching  ...  Acknowledgements: We are grateful for financial support from RCUK, EPSRC, the Royal Academy of Engineering, and ONR MURI N00014-07-1-0182.  ... 
doi:10.1109/cvprw.2009.5206523 fatcat:drop7dorfffenlbevbswxgxj6y

Learning sign language by watching TV (using weakly aligned subtitles)

Patrick Buehler, Andrew Zisserman, Mark Everingham
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
; (ii) we show that by optimizing a scoring function based on multiple instance learning, we are able to extract the sign of interest from hours of signing footage, despite the very weak and noisy supervision  ...  The contributions are: (i) we propose a distance function to match signing sequences which includes the trajectory of both hands, the hand shape and orientation, and properly models the case of hands touching  ...  Acknowledgements: We are grateful for financial support from RCUK, EPSRC, the Royal Academy of Engineering, and ONR MURI N00014-07-1-0182.  ... 
doi:10.1109/cvpr.2009.5206523 dblp:conf/cvpr/BuehlerZE09 fatcat:3ya3rjsf3zafvppf7njx2vxpdm

PyTorch Connectomics: A Scalable and Flexible Segmentation Framework for EM Connectomics [article]

Zudi Lin, Donglai Wei, Jeff Lichtman, Hanspeter Pfister
2021 arXiv   pre-print
We present PyTorch Connectomics (PyTC), an open-source deep-learning framework for the semantic and instance segmentation of volumetric microscopy images, built upon PyTorch.  ...  Those functionalities can be easily realized in PyTC by changing the configuration options without coding and adapted to other 2D and 3D segmentation tasks for different tissues and imaging modalities.  ...  model that additionally learns a signed distance map (Fig. 4 ).  ... 
arXiv:2112.05754v1 fatcat:jjdatdvbr5eypmxmqbvujmarc4

Recurrent Pixel Embedding for Instance Grouping [article]

Shu Kong, Charless Fowlkes
2017 arXiv   pre-print
We introduce a differentiable, end-to-end trainable framework for solving pixel-level grouping problems such as instance segmentation consisting of two novel components.  ...  We demonstrate substantial improvements over state-of-the-art instance segmentation for object proposal generation, as well as demonstrating the benefits of grouping loss for classification tasks such  ...  Rahul Sukthankar for the helpful discussion, advice and encouragement.  ... 
arXiv:1712.08273v1 fatcat:77ohxblx3vgpjnwr5loalgqeam

A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images

Furong Shi, Tong Zhang
2021 Remote Sensing  
Based on the multi-scale features, one regression loss and two classification losses were used for predicting the distance-transform map, segmentation, and boundary.  ...  In order to compensate for the loss of shape information, two shape-related auxiliary tasks (i.e., boundary prediction and distance estimation) were jointly learned with building segmentation task in our  ...  Distance representations can also be used to supplement shape information for semantic segmentation. In [34] , a signed distance representation was introduced for building extraction.  ... 
doi:10.3390/rs13142656 fatcat:5lfaohfky5bmdhchqfbjvzmrcm

Recurrent Pixel Embedding for Instance Grouping

Shu Kong, Charless Fowlkes
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We introduce a differentiable, end-to-end trainable framework for solving pixel-level grouping problems such as instance segmentation consisting of two novel components.  ...  We demonstrate substantial improvements over state-of-the-art instance segmentation for object proposal generation, as well as demonstrating the benefits of grouping loss for classification tasks such  ...  Fig. 11 shows the embedding visualization, as well as predicted semantic segmentation and instance-level segmentation.  ... 
doi:10.1109/cvpr.2018.00940 dblp:conf/cvpr/KongF18a fatcat:v5rgn7rtl5bhdpggrdzetkm7e4

Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation [article]

Balamurali Murugesan, Kaushik Sarveswaran, Sharath M Shankaranarayana, Keerthi Ram, Jayaraj Joseph, Mohanasankar Sivaprakasam
2019 arXiv   pre-print
For the task of medical image segmentation, fully convolutional network (FCN) based architectures have been extensively used with various modifications.  ...  can be obtained from ground truth segmentation maps with no additional annotation costs.  ...  Distance map D3 is obtained by applying signed distance transform to the contour.  ... 
arXiv:1908.05311v1 fatcat:s2btvguw5rcmha6p37rhog4q7q

ELLIPSDF: Joint Object Pose and Shape Optimization with a Bi-level Ellipsoid and Signed Distance Function Description [article]

Mo Shan, Qiaojun Feng, You-Yi Jau, Nikolay Atanasov
2021 arXiv   pre-print
This paper proposes an expressive yet compact model for joint object pose and shape optimization, and an associated optimization algorithm to infer an object-level map from multi-view RGB-D camera observations  ...  Acknowledgments The first author would like to thank Kejie Li at University of Adelaide for helpful discussions.  ...  The coarse-level shape error function e φ (x, d, T, δz) is defined similarly, using a signed distance function for the coarse shape.  ... 
arXiv:2108.00355v1 fatcat:ufmebg4v6rd63c45ty67au7pte

DeepCSR: A 3D Deep Learning Approach for Cortical Surface Reconstruction [article]

Rodrigo Santa Cruz, Leo Lebrat, Pierrick Bourgeat, Clinton Fookes, Jurgen Fripp, Olivier Salvado
2020 arXiv   pre-print
Towards this end, we train a neural network model with hypercolumn features to predict implicit surface representations for points in a brain template space.  ...  Traditional frameworks for this task like FreeSurfer demand lengthy runtimes, while its accelerated variant FastSurfer still relies on a voxel-wise segmentation which is limited by its resolution to capture  ...  For a given surface S, the function f S can be modeled, for instance, using occupancy field or signed distance function.  ... 
arXiv:2010.11423v1 fatcat:yuiaqrjx75gexcl3vhghblhfi4

Driving Scene Perception Network: Real-Time Joint Detection, Depth Estimation and Semantic Segmentation

Liangfu Chen, Zeng Yang, Jianjun Ma, Zheng Luo
2018 2018 IEEE Winter Conference on Applications of Computer Vision (WACV)  
As the demand for enabling high-level autonomous driving has increased in recent years and visual perception is one of the critical features to enable fully autonomous driving, in this paper, we introduce  ...  an efficient approach for simultaneous object detection, depth estimation and pixel-level semantic segmentation using a shared convolutional architecture.  ...  In our preliminary experiment, without limiting the distance range of objects for detection at training stage, the network falls to detecting many imaginary traffic signs.  ... 
doi:10.1109/wacv.2018.00145 dblp:conf/wacv/ChenYML18 fatcat:rvsk22hyz5axrmrsppfgijebjm

Table of Contents

2021 2021 Swedish Artificial Intelligence Society Workshop (SAIS)  
Signed Distance Functions for Visual Instance Segmentation Emil Brissman, Joakim Johnander, Michael Felsberg 5-10 3 Robot First Aid: Autonomous Vehicles Could Help in Emergencies Martin Cooney,  ...  Drill Core Analysis Christian Günther, Nils Jansson, Marcus Liwicki, Foteini Simistira-Liwicki 19-24 6 Class-Incremental Learning for Semantic Segmentation -A study Karl Holmquist, Lena Klasén,  ... 
doi:10.1109/sais53221.2021.9483990 fatcat:fk7g3pkivzcdpc56rimeyv6csq

A simple model of the vertical–horizontal illusion

Pascal Mamassian, Marie de Montalembert
2010 Vision Research  
In particular, we find that the '+'-sign figure suffers from a loss of sensitivity in comparing their vertical and horizontal segments when compared to the 'L'-figure.  ...  These two factors, orientation anisotropy and length bisection, provide a very good account of various configurations of the illusion when the stimulus looks like a 'T', an 'L', or a '+'-sign, and for  ...  Acknowledgments The authors thank Michael Landy and Peter Thompson for their comments on an earlier version of this manuscript.  ... 
doi:10.1016/j.visres.2010.03.005 pmid:20298713 fatcat:lsqkahaufnapvkfleqpnsc535y

Learning Metric Graphs for Neuron Segmentation In Electron Microscopy Images [article]

Kyle Luther, H. Sebastian Seung
2019 arXiv   pre-print
We first show that seed-based postprocessing of the feature vectors, as originally proposed, produces inferior accuracy because it is difficult for the convolutional net to predict feature vectors that  ...  In this case, segmentations from a "metric graph" turn out to be competitive or even superior to segmentations from a directly predicted affinity graph.  ...  We can derive an affinity graph from a metric graph by simply inverting the signs of all the distances between objects (i.e. affinity is the negative distance between nodes).  ... 
arXiv:1902.00100v1 fatcat:35mbkttjxja7jaqabykc3tlkxu

Latent Partition Implicit with Surface Codes for 3D Representation [article]

Chao Chen, Yu-Shen Liu, Zhizhong Han
2022 arXiv   pre-print
LPI represents a shape as Signed Distance Functions (SDFs) using surface codes.  ...  LPI can be learned without ground truth signed distances, point normals or any supervision for part partition.  ...  Visual comparison with DeepLS [8] , SIF [22] and Nglod [80] under D-FAUST. Fig. 12 . 12 Fig. 12. Shape abstraction with instance segmentation.  ... 
arXiv:2207.08631v3 fatcat:pff5z6ruo5g3rne46apro7z7hu
« Previous Showing results 1 — 15 out of 68,294 results