A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Ordered or Orderless: A Revisit for Video based Person Re-Identification
[article]
2019
arXiv
pre-print
Preserved Fulltext
Is recurrent network really necessary for learning a good visual representation for video based person re-identification (VPRe-id)? ...
Our work also presents a promising way to bridge the gap between video and image based person re-identification. ...
In this paper, we consider more practical scenarios of video based person re-identification (VPRe-id), in which a video of a person, as seen in one camera, must be matched against a gallery of videos captured ...
arXiv:1912.11236v1
fatcat:zwjzjg2r35epzohu3wlzcul2g4
2021 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 43
2022
IEEE Transactions on Pattern Analysis and Machine Intelligence
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
The Author Index contains the primary entry for each item, listed under the first author's name. ...
-that appeared in this periodical during 2021, and items from previous years that were commented upon or corrected in 2021. ...
., +, TPAMI Feb. 2021 745-752 Ordered or Orderless: A Revisit for Video Based Person Re-Identification. ...
doi:10.1109/tpami.2021.3126216
fatcat:h6bdbf2tdngefjgj76cudpoyia
Noise-Resistant Deep Metric Learning with Probabilistic Instance Filtering
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we bridge this important gap by proposing Probabilistic Ranking-based Instance Selection with Memory (PRISM) approach for DML. ...
Specifically, we propose a novel method, namely the von Mises-Fisher Distribution Similarity (vMF-Sim), to calculate this probability by estimating a von Mises-Fisher (vMF) distribution for each data class ...
Shen, “Ordered or orderless: A revisit for video based person 2009, vol. 494.
re-identification,” IEEE transactions on pattern analysis and machine [47] J. Krause, M. Stark, J. ...
arXiv:2108.01431v2
fatcat:vz5d4iqshvds7iaqohl7js4ieq
Multiple Object Tracking: A Literature Review
[article]
2017
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We inspect the recent advances in various aspects and propose some interesting directions for future research. ...
We endeavor to provide a thorough review on the development of this problem in recent decades. ...
Possible combinations include object segmentation [174] , human re-identification [175] , human pose estimation [17] , and action recognition [18] . ...
arXiv:1409.7618v4
fatcat:ff544dkiibfvzjvdz6gellsjh4
Table of contents
2021
ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Han, Xin Tian, Yuhong Yang, Wuhan University, China MMSP-5.3: A TRIPLET APPEARANCE PARSING NETWORK FOR PERSON ................................................ 4245 RE-IDENTIFICATION Mingfu Xiong, Engineering ...
MODEL FOR PERSON RE-IDENTIFICATION ............................... 1495 Xipeng Lin, Yubin Yang, Nanjing University, China IVMSP-17.4: CROWD COUNTING VIA MULTI-LEVEL REGRESSION WITH LATENT ............ ...
doi:10.1109/icassp39728.2021.9414617
fatcat:m5ugnnuk7nacbd6jr6gv2lsfby
Image Retrieval: Modelling Keywords via Low-level Features
2015
ELCVIA Electronic Letters on Computer Vision and Image Analysis
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
In recent years, much effort has been invested on automatic image annotation methods, since the manual assignment of keywords (which is necessary for text-based image retrieval) is a time consuming and ...
It discusses and presents several studies referring to: (a) low-level feature extraction and selection for the task of automatic annotation of images, (b) training algorithms that can be utilized for keyword ...
Thereby, a small group of experts is necessary for a high quality video labeling. ...
doi:10.5565/rev/elcvia.725
fatcat:zbpgc3jstnd5nf5ltcly3rdbhi
Drawing a Line in the Patent Subject Matter Sands: Does Europe Provide a Solution to the Business Method and Software Patent Problem?
2010
Social Science Research Network
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2012; you can also visit the original URL.
The file type is application/pdf.
are doing so in order to promote order or safety.103
III. ...
In 1968, in In re Prater, the CCPA considered a claim rejection based on In re Abrams' "three rules" for dealing with mental steps.238 The court noted that the defense attorney's brief proposed the three ...
Introduction On February 24, 2010, Italian Judge Oscar Magi convicted three Google executives for violations of Italian data protection laws after a video was uploaded to Google Video showing an autistic ...
doi:10.2139/ssrn.1676898
fatcat:hw2tqdw5n5aflm26bz3cfquj6a
Recovery of stereopsis in human adults with strabismus through perceptual learning
2010
Journal of Vision
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2012; you can also visit the original URL.
The file type is application/pdf.
Similarly, when a human observes a short video, it is easy to decide if the event taking place in the video is normal or unexpected-even if the video depicts an unfamiliar place for the viewer. ...
a hump) necessary for object identification. ...
We set up an exhibition with 6 high-quality canvas prints and installed a radio-based identification (RFID) system to track 60 single visitors. ...
doi:10.1167/10.7.1124
fatcat:w3a4vu2hybhcvjwfqrryzzilbi
Machine Learning-Based Routing and Wavelength Assignment in Software-Defined Optical Networks
2019
IEEE Transactions on Network and Service Management
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Then, in Chapter 7 we revisit meta-information of Android applications to attempt the identification of groups of potentially repackaged applications using qualitative fields, such as application title ...
Then, cosine similarity is computed over each pair or application descriptions' BoW to produce an orderless and language-independent similarity score for application descriptions. ...
doi:10.1109/tnsm.2019.2927867
fatcat:or3lhqdqbnas3cztwkqr5ykhhq
Area Chairs
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Only the very highest quality papers were selected for oral presentation, with 38 papers gaining a podium spot, or 10% of the submissions. ...
Of the 365 submissions, just 144 were accepted for presentation in BMVC 2016, which is a 39% acceptance rate. ...
We notice that most errors (around or above 90% of them) are below 2.5 meters, that is below 3 pixels for the image resolution available in our experiments. ...
fatcat:qy4idlefnzf6xjwaxjjpd75tzy
Deep Zero- and Few-shot Learning in Computer Vision
[article]
2020
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
is from a seen or unseen class. ...
Zero-shot learning requires learning a mapping that associates feature vectors extracted from images with semantic annotations that describe objects and/or scenes of interest. ...
For instance, we cannot have thousands of photos of every person to get a robust model in surveillance videos for security checks. ...
doi:10.25911/5f5a020a5c26f
fatcat:7etnry26vnbqbg752j77ikirwm
Reframing the World. Local Communities and Multinational Corporations. The case of Cajamarca, Colombia
2014
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In these statements a concern with the land and nature is clear, but the argument does not describe personal loss, personal endangering or personal connection with the land they defend. ...
For the mining corporation water is a 'resource' that they should
manage in order to guarantee water for people. ...
doi:10.7939/r3h70880t
fatcat:yqmip5c5arhnvltnf3la3u35eq
Object Detection with Active Sample Harvesting
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
The tree is used to select the next sample/image for training, by applying a sampling policy, and the "importance" weights are updated accordingly, to bias the sampling towards informative samples/images ...
The common thread of this dissertation is how to efficiently find the informative samples/images for training. ...
Acknowledgements I would like to express my gratitude to my advisor François Fleuret for his valuable advice all along these four years as well as for always encouraging me to do sound work. ...
doi:10.5075/epfl-thesis-7312
fatcat:nhao2skvyrfyrn2qedzcwwznuu
Opening the machine learning black box with Layer-wise Relevance Propagation
[article]
2019
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
However, due to the nested complex and non-linear structure of many machine learning models, this comes with the disadvantage of them acting as a black box, providing little or no information about the ...
In this thesis, we describe a novel method for explaining non-linear classifier decisions by decomposing the prediction function, called Layer-wise Relevance Propagation (LRP). ...
Acknowledgements
Experimental Setup We demonstrate the results on a classifier for the MIT Places data set (Zhou et al., 2014b) provided by the authors of this data set and the Caffe reference model ...
doi:10.14279/depositonce-7942
fatcat:2ca4ccz5bzb2lojrqrpsxhmzve
PROCEEDINGS OF THE 15TH PYTHON IN SCIENCE CONFERENCE SCHOLARSHIP RECIPIENTS JUMP TRADING AND NUMFOCUS DIVERSITY SCHOLARSHIP RECIPIENTS
Proceedings of the 15th Python in Science Conference July 10-July 16 •
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We are grateful for computer time provided for this purpose by the JARA-HPC We would like to thank Dr. ...
Francesco Pontiggia for helping us solidify many of our data handling and computing ideas and Dr. Jian-You Lin for being the first independent tester of our toolchain. ...
data with ease on a shared computing resource or on personal workstations. ...
fatcat:pkgs32mdl5d3tbfa5xkveedypq
« Previous
Showing results 1 — 15 out of 19 results