Filters








260 Hits in 2.4 sec

Joint Forecasting of Panoptic Segmentations with Difference Attention [article]

Colin Graber, Cyril Jazra, Wenjie Luo, Liangyan Gui, Alexander Schwing
2022 arXiv   pre-print
Because of this, we attain state-of-the-art on panoptic segmentation forecasting metrics.  ...  To address both issues, we study a new panoptic segmentation forecasting model that jointly forecasts all object instances in a scene using a transformer model based on 'difference attention.'  ...  Our panoptic segmentation forecasting.  ... 
arXiv:2204.07157v1 fatcat:g4l3n5vhyrasnlvbavgifjdf5i

Dense Semantic Forecasting in Video by Joint Regression of Features and Feature Motion [article]

Josip Šarić and Sacha Vražić and Siniša Šegvić
2021 arXiv   pre-print
We perform experiments on three dense prediction tasks: semantic segmentation, instance-level segmentation, and panoptic segmentation.  ...  The results reveal state-of-the-art forecasting accuracy across three dense prediction tasks.  ...  Panoptic segmentation forecasting on Cityscapes Table III presents our performance on the panoptic segmentation task.  ... 
arXiv:2101.10777v2 fatcat:t2375opewvbahf3fwtoevmbmuq

Sequential Point Clouds: A Survey [article]

Haiyan Wang, Yingli Tian
2022 arXiv   pre-print
and point cloud forecasting.  ...  This paper presents an extensive review of the deep learning-based methods for sequential point cloud research including dynamic flow estimation, object detection \& tracking, point cloud segmentation,  ...  , panoptic segmentation.  ... 
arXiv:2204.09337v2 fatcat:su7fdqd3xja53jg4zikqcpvoaa

Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches [article]

Omar Elharrouss, Younes Akbari, Noor Almaadeed, Somaya Al-Maadeed
2022 arXiv   pre-print
Panoptic segmentation Panoptic segmentation is a new direction in image segmentation also is the developed version of the instance and semantic segmentation.  ...  In [138] , video panoptic segmentation (VPSnet), which is a new video extension of panoptic segmentation is introduced, in which two types of video panoptic datasets have been used.  ... 
arXiv:2206.08016v1 fatcat:qxfezmrhsvgj7jmazbh5uctlum

Editorial - Marketing and the Rise of Commercial Consumer Surveillance

Jason Pridmore, Detlev Zwick
2011 Surveillance & Society  
aimed at forecasting future positions "in an increasingly dispersed and automated infoscape" (Elmer 2004, 44 ; see also Bogard 1996) .  ...  Hence Elmer (2004) puts the spotlight of contemporary market surveillance strategies on the reproductive (or cybernetic) aspect of electronic panopticism.  ... 
doi:10.24908/ss.v8i3.4163 fatcat:dfivvjhsrveynl6aizw7cbdvei

Snipper: A Spatiotemporal Transformer for Simultaneous Multi-Person 3D Pose Estimation Tracking and Forecasting on a Video Snippet [article]

Shihao Zou, Yuanlu Xu, Chao Li, Lingni Ma, Li Cheng, Minh Vo
2022 arXiv   pre-print
Among these three tasks, pose estimation and tracking are correlated, and tracking is crucial to motion forecasting.  ...  Multi-person pose understanding from RGB videos includes three complex tasks: pose estimation, tracking and motion forecasting.  ...  We show 3D pose tracking and motion forecasting results on JTA and CMU Panoptic datasets, and 2D pose tracking results on Posetrack dataset.  ... 
arXiv:2207.04320v2 fatcat:o3jhrppla5eenfshcx3cbjjmde

Vision-based Uneven BEV Representation Learning with Polar Rasterization and Surface Estimation [article]

Zhi Liu, Shaoyu Chen, Xiaojie Guo, Xinggang Wang, Tianheng Cheng, Hongmei Zhu, Qian Zhang, Wenyu Liu, Yi Zhang
2022 arXiv   pre-print
PolarBEV keeps real-time inference speed on a single 2080Ti GPU, and outperforms other methods for both BEV semantic segmentation and BEV instance segmentation.  ...  (RQ), Segmentation Quality (SQ) and Panoptic Quality (PQ) for evaluating vehicle BEV instance segmentation performance.Our method significantly outperforms FIERY[2]in both performance and speed.  ...  Instead of IoU metric, we also report Panoptic Quality (PQ) metric for evaluating vehicle BEV instance segmentation performance, following previous work [2] .  ... 
arXiv:2207.01878v1 fatcat:huwtwmjltnea3pkwqdgqq46aee

Predicting what you'll do tomorrow: Panspectric surveillance and the contemporary corporation

Karl Palmås
2010 Surveillance & Society  
In contrast to the original panoptic architectures, the panspectron monitors a wider segment of frequencies of the electromagnetic spectrum.  ...  Panoptic business and panspectric business compared As hinted above, panspectrocism should -like panopticism -be seen as a new source of scale economies.  ... 
doi:10.24908/ss.v8i3.4168 fatcat:bchmr5dz4ratbljb7d5v4zwpui

How2Sign: A Large-scale Multimodal Dataset for Continuous American Sign Language

Amanda Duarte, Shruti Palaskar, Lucas Ventura, Deepti Ghadiyaram, Kenneth DeHaan, Florian Metze, Jordi Torres, Xavier Giro-i-Nieto
2021 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Gloss Annotation HI, ME FS-AMELIA WILL ME TALK GUM IX-LOC-HAIR STUCK Green screen studio RGB videos Green screen studio RGB-D videos Body-face-hands keypoints Panoptic studio data (only for a subset) 3D  ...  Panoptic studio. The Panoptic studio [17] is a system equipped with 480 VGA cameras, 30 HD cameras and 10 RGB-D sensors, all synchronized.  ...  ., where the data needs to be segmented and annotated at the sentence level. Currently, there is no ASL dataset large enough to be used with recent deep learning approaches.  ... 
doi:10.1109/cvpr46437.2021.00276 fatcat:ppgq7aypbzhibp5rojpgifods4

Evaluation of Deep Learning-Based Neural Network Methods for Cloud Detection and Segmentation

Stefan Hensel, Marin B. Marinov, Michael Koch, Dimitar Arnaudov
2021 Energies  
These were used to train several state-of-the-art deep neural networks for object detection and segmentation.  ...  Deep neural networks, based on the two-stage R-CNN architecture, were trained and compared with a U-net segmentation approach implemented by CloudSegNet.  ...  The structure is highly modular, allowing networks to be adapted and trained for detection and segmentation, the latter as a classic instance, or for panoptic variation.  ... 
doi:10.3390/en14196156 fatcat:fiqkqekdjreg3niu5p3jtruaeq

Digital Technology and the Market for Political Surveillance

Philip N. Howard, John N. Carr, Tema J. Milstein
2002 Surveillance & Society  
addresses the manner in which such technological innovations have enabled a dramatically expanded market for public policy opinion data, and explores the potential role of that market in facilitating panoptic  ...  Accordingly, the segmentation and division of the polis through techniques of dataveillance can "create and exacerbate inequalities that can distort public discussion and debate" (Danna & Gandy, 2002;  ...  given the current aggressive efforts by the United States to export and e xpand the concomitant ideologies of open markets and 'democracy' (Harvey, 2003; Panitch & Gindin, 2003) , we may quite easily forecast  ... 
doi:10.24908/ss.v3i1.3320 fatcat:ywoeurqvc5f6tbz33h4nwmjkci

How2Sign: A Large-scale Multimodal Dataset for Continuous American Sign Language [article]

Amanda Duarte, Shruti Palaskar, Lucas Ventura, Deepti Ghadiyaram, Kenneth DeHaan, Florian Metze, Jordi Torres, Xavier Giro-i-Nieto
2021 arXiv   pre-print
A three-hour subset was further recorded in the Panoptic studio enabling detailed 3D pose estimation.  ...  Panoptic studio. The Panoptic studio [17] is a system equipped with 480 VGA cameras, 30 HD cameras and 10 RGB-D sensors, all synchronized.  ...  ., where the data needs to be segmented and annotated at the sentence level. Currently, there is no ASL dataset large enough to be used with recent deep learning approaches.  ... 
arXiv:2008.08143v2 fatcat:cfmuxp5dgjdipdrt2i4ep2hcry

Cotton Leaf Disease Detection Using Instance Segmentation

Prashant Udawant, Pravin Srinath
2022 Journal of Cases on Information Technology  
Detectron 2 supports object detection, semantic segmentation, instance segmentation, panoptic segmentation, and keypoint detection.  ...  ., 2014) that the disease be controlled and damage reduced by measuring plant leaf damage levels and thereby forecasting plant damage levels.  ... 
doi:10.4018/jcit.296721 fatcat:pupi6nmvsbdvpncd53hocyj6y4

The Culture of Surveillance: G-Men Redux and Total Information Awareness

Leslie Regan Shade
2003 TOPIA Canadian Journal of Cultural Studies  
ICTs are essential not only for creating the panoptic sort, but also for "the analytic approaches that differentiate, classify, segment, and target individuals and groups upon the basis of models, assumptions  ...  Although Gandy's panoptic sort is controlled by consumer information collected and controlled by the capitalist marketplace, we could consider TIA to be a hyper-panoptic sort, as it relies on both the  ... 
doi:10.3138/topia.9.35 fatcat:vcycpvskirgpbfxhwhw7pph3aq

RID—Roof Information Dataset for Computer Vision-Based Photovoltaic Potential Assessment

Sebastian Krapf, Lukas Bogenrieder, Fabian Netzler, Georg Balke, Markus Lienkamp
2022 Remote Sensing  
Therefore, we present the Roof Information Dataset for semantic segmentation of roof segments and roof superstructures.  ...  By providing the datasets in initial and reviewed versions, we promote a data-centric approach for the semantic segmentation of roof information.  ...  Implementation details for U-Net and Panoptic FPN.  ... 
doi:10.3390/rs14102299 fatcat:t3zgj3abrvbz7l2f3sgmxnuuxm
« Previous Showing results 1 — 15 out of 260 results