Filters








137,590 Hits in 5.3 sec

Estimation of Correspondent Trajectories in Multiple Overlapping Synchronized Videos Using Correlation of Activity Functions

Thierry Malon, Sylvie Chambon, Vincent Charvillat, Alain Crouzil
2019 2019 IEEE International Conference on Image Processing (ICIP)  
The ranking depends on how they allow to visualize as best as possible, i.e. with significant details, a trajectory query drawn in one of the videos.  ...  We present an approach for ranking a collection of videos with overlapping fields of view.  ...  Figure 2 presents the top 3 best views proposed for different trajectory queries drawn in different views.  ... 
doi:10.1109/icip.2019.8803032 dblp:conf/icip/MalonCCC19 fatcat:vui3f6katraethbdjrhq4krlde

Shared Cross-Modal Trajectory Prediction for Autonomous Driving [article]

Chiho Choi, Joon Hee Choi, Srikanth Malla, Jiachen Li
2021 arXiv   pre-print
Predicting future trajectories of traffic agents in highly interactive environments is an essential and challenging problem for the safe operation of autonomous driving systems.  ...  ., LiDAR data) is required to generate predictions from the input perspective (i.e., in the LiDAR space), while taking advantages from the model trained with multiple sensor modalities.  ...  Note that existing works solve the problem either in top-down view [23, 36, 9] with LiDAR data or in frontal view [46, 4, 30] with RGB images.  ... 
arXiv:2004.00202v3 fatcat:hlo5wxhctvac7kvjjk6wd4u74m

Shared Cross-Modal Trajectory Prediction for Autonomous Driving [article]

Chiho Choi, Joon Hee Choi, Jiachen Li, Srikanth Malla
2021 arXiv   pre-print
Predicting future trajectories of traffic agents in highly interactive environments is an essential and challenging problem for the safe operation of autonomous driving systems.  ...  ., LiDAR data) is required to generate predictions from the input perspective (i.e., in the LiDAR space), while taking advantages from the model trained with multiple sensor modalities.  ...  Note that existing works solve the problem either in top-down view [23, 36, 9] with LiDAR data or in frontal view [46, 4, 30] with RGB images.  ... 
arXiv:2011.08436v2 fatcat:g7gj34v2szfahokrwhcqlmwkuy

Revisiting revisits in trajectory recommendation [article]

Aditya Krishna Menon, Dawei Chen, Lexing Xie, Cheng Soon Ong
2017 arXiv   pre-print
Trajectory recommendation is the problem of recommending a sequence of places in a city for a tourist to visit.  ...  Third, we conduct experiments on real-world trajectory recommendation datasets to identify the tradeoffs imposed by each of the three approaches.  ...  To find the best sequence without loops in Equation 2, one can apply a list Viterbi algorithm to find not just the single best sequence, but rather the top K best sequences.  ... 
arXiv:1708.05165v1 fatcat:mt72pvivefhateqoeeaq35hcx4

On Exposing the Challenging Long Tail in Future Prediction of Traffic Actors [article]

Osama Makansi, Özgün Cicek, Yassine Marrakchi, Thomas Brox
2021 arXiv   pre-print
As a consequence, we propose to supplement the usual loss with aloss that places challenging cases closer to each other.  ...  Code is available at https://github.com/lmb-freiburg/Contrastive-Future-Trajectory-Prediction  ...  For the LDAM baseline [5] , we experiment with different scaling factors and use the best setting s = 1.  ... 
arXiv:2103.12474v3 fatcat:53pvzeuatng7tcgtf6kmiycphy

Identifying Same Persons from Temporally Synchronized Videos Taken by Multiple Wearable Cameras

Kang Zheng, Hao Guo, Xiaochuan Fan, Hongkai Yu, Song Wang
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
In this paper, we study cross-video person identification for wearable cameras, which are constantly moving with the wearers.  ...  ., determining whether at a given time, persons tracked in different videos are the same person or not, is a key step to integrate such multi-view information for collaborative action recognition.  ...  We then remove this best matched 2D mocap sequence from Z and apply Eq. (4) again to find the second best matched 2D mocap sequence. To obtain top-K best matched ones, we repeat this process K times.  ... 
doi:10.1109/cvprw.2016.106 dblp:conf/cvpr/ZhengGFYW16 fatcat:ozhr3h75gvgorb7rwrrxwpqmae

Scene Compliant Trajectory Forecast with Agent-Centric Spatio-Temporal Grids [article]

Daniela Ridel, Nachiket Deo, Denis Wolf, Mohan Trivedi
2019 arXiv   pre-print
We represent the past trajectories of agents using binary 2-D grids, and the underlying scene as a RGB birds-eye view (BEV) image, with an agent-centric frame of reference.  ...  Results on the publicly available Stanford Drone Dataset (SDD) show that our model outperforms prior approaches and outputs realistic future trajectories that comply with scene structure and past motion  ...  ACKNOWLEDGMENT The authors are grateful for the generous and continuing support of the research by sponsors and team members of the UCSD LISA and CVRR labs.  ... 
arXiv:1909.07507v1 fatcat:qezp5et63jgmleks6766jdo22e

Structured Recommendation [article]

Dawei Chen, Lexing Xie, Aditya Krishna Menon, Cheng Soon Ong
2017 arXiv   pre-print
Motivated by trajectory recommendation, we focus on sequential structures but in contrast to classical Viterbi decoding we require that valid predictions are sequences with no repeated elements.  ...  Experiments on real-world trajectory recommendation datasets show the benefits of our approach over existing, non-structured recommendation approaches.  ...  We report results on the best of top k [42] .  ... 
arXiv:1706.09067v1 fatcat:e27aao47szhvlgr6u4lz6jmmq4

Three-view uncalibrated visual servoing

Azad Shademan, Martin Jägersand
2010 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems  
Simulations with a 6-DOF eye-in-hand manipulator show that the proposed uncalibrated approach rapidly converges in all cases.  ...  The visual servoing performance is evaluated for various easy and challenging motions, such as the translational motion along the view axis and the large rotation around the view axis.  ...  The camera trajectory is almost linear with control parameter β is 0.5 (other values have similar performance). Fig. 4 (top-right) shows the image trajectories.  ... 
doi:10.1109/iros.2010.5649988 dblp:conf/iros/ShademanJ10 fatcat:ni3evbx66rcojmjszvu53fhrmy

Mining GPS Data for Trajectory Recommendation [chapter]

Peifeng Yin, Mao Ye, Wang-Chien Lee, Zhenhui Li
2014 Lecture Notes in Computer Science  
The wide use of GPS sensors in smart phones encourages people to record their personal trajectories and share them with others in the Internet.  ...  The Hybrid recommendation combines the ABR and GBR. We finally conducted extensive experiments to evaluate these proposed solutions and it turned out the hybrid solution displays the best performance.  ...  For GPS feature, it is shown that the combination of partial-view feature and entire-view feature achieves best performance.  ... 
doi:10.1007/978-3-319-06605-9_5 fatcat:f2nwvmx3vzcnrmp2hm6vtv6svy

SimAug: Learning Robust Representations from Simulation for Trajectory Prediction [article]

Junwei Liang, Lu Jiang, Alexander Hauptmann
2020 arXiv   pre-print
The key idea is to mix the feature of the hardest camera view with the adversarial feature of the original view. We refer to our method as SimAug.  ...  This paper studies the problem of predicting future trajectories of people in unseen cameras of novel scenarios and views.  ...  The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA  ... 
arXiv:2004.02022v3 fatcat:xvjrrrmvsrgyxdojcjrrebs5ai

Personalized Trajectory Prediction via Distribution Discrimination [article]

Guangyi Chen, Junlong Li, Nuoxing Zhou, Liangliang Ren, Jiwen Lu
2021 arXiv   pre-print
Trajectory prediction is confronted with the dilemma to capture the multi-modal nature of future dynamics with both diversity and accuracy.  ...  Our method can be integrated with existing multi-modal stochastic predictive models as a plug-and-play module to learn the more discriminative latent distribution.  ...  , e.g. we calculate the minimum ADE/FDE of top 20 probability trajectories, when k = 20 M .  ... 
arXiv:2107.14204v1 fatcat:xgmwe54qb5c7xhrzwk2evor74e

Directional Beams of Dense Trajectories for Dynamic Texture Recognition [chapter]

Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frédéric Bouchara, Xuan Son Nguyen
2018 Lecture Notes in Computer Science  
Furthermore, we also exploit motion information from dense trajectories to boost the discriminative power of the proposed descriptor.  ...  a new operator, named LVP f ull -TOP, based on local vector patterns (LVP) in full-direction on three orthogonal planes.  ...  Fig. 2 . 2 (Best viewed in color) Encoding FDT with the proposed LVP f ull -TOP in direction local vector pattern of a pixel q i at the frame f i based on full-direction in three orthogonal planes; p  ... 
doi:10.1007/978-3-030-01449-0_7 fatcat:ho5v3iiw5fg2lhmdugvlnpqlre

The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction [article]

Junwei Liang, Lu Jiang, Kevin Murphy, Ting Yu, Alexander Hauptmann
2020 arXiv   pre-print
We show that our model achieves the best results on our dataset, as well as on the real-world VIRAT/ActEV dataset (which just contains one possible future).  ...  This provides the first benchmark for quantitative evaluation of the models to predict multi-future trajectories.  ...  The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA  ... 
arXiv:1912.06445v3 fatcat:luumpu6qmfeptndjdxqsexfkve

Trajectory Fusion for Multiple Camera Tracking [chapter]

Ariel Amato, Murad Haj, Mikhail Mozerov, Jordi Gonzàlez
2007 Advances in Soft Computing  
The proposed approach is based on the matching of multiple trajectories from multiple views using spatial and temporal information.  ...  The proposed method can also be considered as an interpolation algorithm of a disconnected trajectory during the time of occlusion.  ...  Acknowledgements This work is supported by EC grants IST-027110 for the HERMES project and IST-045547 for the VIDI-video project, and by the Spanish MEC under  ... 
doi:10.1007/978-3-540-75175-5_3 dblp:series/asc/AmatoHMG08 fatcat:hcxug5vcszfwvgnnkrjcsrovo4
« Previous Showing results 1 — 15 out of 137,590 results