A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2011; you can also visit the original URL.
The file type is application/pdf.
Filters
IFS-CoCo: Instance and feature selection based on cooperative coevolution with nearest neighbor rule
2010
Pattern Recognition
Preserved Fulltext
The proposed approach is compared with a wide range of evolutionary feature and instance selection methods for classification. ...
The results contrasted through non-parametric statistical tests show that our model outperforms previously proposed evolutionary approaches for performing data reduction processes in combination with the ...
Selection of suitable parameters for IFS-CoCo Although we have defined completely how IFS-CoCo works, an interesting question remains: How can a user select suitable values for the parameters of the algorithm ...
doi:10.1016/j.patcog.2009.12.012
fatcat:5pv2df7acff7npsifpzfhymhn4
Proteomics of E.coli Nissle 1917 in Responce to Cocos nucifera sap and Wine
2015
International Letters of Natural Sciences
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 2022; you can also visit the original URL.
The file type is application/pdf.
Table -17
: Recommended Search parameters for search MALDI-MS and MS/MS data. ...
For Colloidal coomassie blue detection, gels were fixed for 1 h in 20% ethanol 7% acetic acid then wash gel in water 3 times for 10 minutes each. ...
Small toxic polypeptide LDRA_ECOLi are to be undergoing for drug designing which will be useful for bio-pharmaceutical industries to prepare the drugs against the toxicity of wine treatment. ...
doi:10.56431/p-mobj8o
fatcat:5kl2hipnp5g5lgmpg2tvkmhyj4
Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes
[article]
2022
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 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Code is available at https://github.com/tianyu0207/PEBAL. ...
More specifically, PEBAL is based on a non-trivial joint training of EBM and AL, where EBM is trained to output high-energy for anomaly pixels (from outlier exposure) and AL is trained such that these ...
To train the model in (1), we formulate a cost function that jointly trains PAL and EBM to classify anomalous pixels. An important training hyper-parameter Inlier Logits Outlier Logits Seg. ...
arXiv:2111.12264v6
fatcat:dpgqevqzljdlrp5otublhd2ova
Observations of dune interactions from DEMs using through-water Structure from Motion
2020
Geomorphology
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 2022; you can also visit the original URL.
The file type is application/pdf.
It is a low-cost and non-intrusive alternative for submerged bed elevation measurements. ...
New data for 23 dune field DEMs from 6 experiments were obtained and are presented. The dune field DEM time series provide insights on dune interaction processes. ...
Oane Galama is thanked for drawing the experimental setup for the inset in Figure 1 . We acknowledge supportive and constructive reviews by Arjan Reesink and Gerardo Perillo. ...
doi:10.1016/j.geomorph.2020.107126
fatcat:n7ex7vfm4bbgnpt6gbhketmigu
Measuring Non-Gaussianity by Phi-Transformed and Fuzzy Histograms
2012
Advances in Artificial Neural Systems
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.
Independent component analysis (ICA) is an essential building block for data analysis in many applications. ...
We introduce a very general technique for evaluating ICA results rooted in information-theoretic model selection. ...
CoCo [16] , a technique for parameter-free outlier detection, is based on the ideas of data compression and coding costs. ...
doi:10.1155/2012/962105
fatcat:umottlr4f5d23p3rksafp4ztem
Automamatic detection of discontinuities in the station position time series of the reprocessed global GNSS network using Bernese GNSS Software
2020
Acta geodynamica et geomaterialia
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 program is designed for the automatic analysis of time series, in which the functional model is adapted to the time series of coordinates depending on the adopted parameters. ...
The results show that the optimum confidence level for the autonomous detection of station discontinuities in FODITS is 99 % and 98 %, for 7-day and 3-day GNSS solutions, respectively, when compared to ...
The threshold for outliers was set at 5σ, because the CODE solutions have already been pre-filtered at the observation level and should, therefore, be free from gross errors. ...
doi:10.13168/agg.2020.0032
fatcat:vbxqyytq3jdorjxtifett4327y
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
[article]
2016
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 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. ...
In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available. ...
For the feature-shared variant, the result is 69.9%-better than the strong SS baseline, yet with nearly cost-free proposals. ...
arXiv:1506.01497v3
fatcat:dbcpx3w4fzgvxnlblbuy72e3hu
RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free
[article]
2019
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 2020; you can also visit the original URL.
The file type is application/pdf.
The detection component of RetinaMask has the same computational cost as the original RetinaNet, but is more accurate. ...
COCO test-dev results are up to 41.4 mAP for RetinaMask-101 vs 39.1mAP for RetinaNet-101, while the runtime is the same during evaluation. ...
Acknowledgements We thank Tamara Berg, and Phil Ammirato for their helpful suggestions, and we acknowledge support from NSF 1452851, 1533771, 1526367. ...
arXiv:1901.03353v1
fatcat:5fxiyqokhzezfmflblo4foejtm
Mask-Free Video Instance Segmentation
[article]
2023
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 2023; you can also visit the original URL.
The file type is application/pdf.
Our mask-free objective is simple to implement, has no trainable parameters, is computationally efficient, yet outperforms baselines employing, e.g., state-of-the-art optical flow to enforce temporal mask ...
Our code and trained models are available at https://github.com/SysCV/MaskFreeVis. ...
Image Mask: using COCO [31] image mask labels for image-based pretraining. Pseudo Video: using Pseudo Videos from COCO images for joint training [58] . ...
arXiv:2303.15904v1
fatcat:5q4jwu4p35gnzcfbwbpiiq6w4i
BRACE: The Breakdancing Competition Dataset for Dance Motion Synthesis
[article]
2022
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 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Generative models for audio-conditioned dance motion synthesis map music features to dance movements. ...
We adopt a hybrid labelling pipeline leveraging deep estimation models as well as manual annotations to obtain good quality keypoint sequences at a reduced cost. ...
The used hyper-parameters are the same as those specified in the official code repository. ...
arXiv:2207.10120v2
fatcat:u2ozlg6u7nbmlimuuhj542amou
Foreground-Background Imbalance Problem in Deep Object Detectors: A Review
[article]
2020
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 2020; you can also visit the original URL.
The file type is application/pdf.
Recent years have witnessed the remarkable developments made by deep learning techniques for object detection, a fundamentally challenging problem of computer vision. ...
Third, we experimentally compare the performance of some state-of-the-art solutions on the COCO benchmark. Promising directions for future work are also discussed. ...
If there are multiple evaluation metrics in the real-world application, we recommend the sampling-free mechanism as a baseline method, whose code can be found at https://github.com/ChenJoya/sampling-free ...
arXiv:2006.09238v1
fatcat:nxb46vw4mrff3ahkdhjydqjmlu
K-NS: Section-Based Outlier Detection in High Dimensional Space
[article]
2014
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 2020; you can also visit the original URL.
The file type is application/pdf.
Our proposed approach not only detects outliers in low dimensional space with section-density ratio but also detects outliers in high dimensional space with the ratio of k-nearest section against average ...
However, as the dimension increases, most of these existing methods perform poorly in detecting outliers because of "high dimensional curse". ...
CoCo: coding cost for parameter-free outlier detection. ...
arXiv:1405.1027v1
fatcat:m66qqqjlwncnje6dfi2dvr5olm
Overcoming Classifier Imbalance for Long-Tail Object Detection With Balanced Group Softmax
2020
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
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 2022; you can also visit the original URL.
The file type is application/pdf.
Code is available at https://github.com/FishYuLi/ BalancedGroupSoftmax. ...
We find existing detection methods are unable to model few-shot classes when the dataset is extremely skewed, which can result in classifier imbalance in terms of parameter magnitude. ...
General object detection Deep learning based object detection frameworks are divided into anchor-based and anchor-free ones. ...
doi:10.1109/cvpr42600.2020.01100
dblp:conf/cvpr/LiWKTWLF20
fatcat:uts5xdndgfhjjfv3kppdrggciy
Joint Detection and Recounting of Abnormal Events by Learning Deep Generic Knowledge
2017
2017 IEEE International Conference on Computer Vision (ICCV)
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.
Our approach first learns CNN with multiple visual tasks to exploit semantic information that is useful for detecting and recounting abnormal events. ...
Our approach outperforms the stateof-the-art on Avenue and UCSD Ped2 benchmarks for abnormal event detection and also produces promising results of abnormal event recounting. ...
While automatic anomaly detection can free people from having to monitor videos, we still have to check videos when the systems raise alerts, and this still involves immense costs. ...
doi:10.1109/iccv.2017.391
dblp:conf/iccv/HinamiMS17
fatcat:hgwrqpuxqnfwxeetg6afjmcqqy
Joint Detection and Recounting of Abnormal Events by Learning Deep Generic Knowledge
[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 2020; you can also visit the original URL.
The file type is application/pdf.
Our approach first learns CNN with multiple visual tasks to exploit semantic information that is useful for detecting and recounting abnormal events. ...
Our approach outperforms the state-of-the-art on Avenue and UCSD Ped2 benchmarks for abnormal event detection and also produces promising results of abnormal event recounting. ...
While automatic anomaly detection can free people from having to monitor videos, we still have to check videos when the systems raise alerts, and this still involves immense costs. ...
arXiv:1709.09121v1
fatcat:ecuzpw6yxvcktgqcx2v4mc4snm
« Previous
Showing results 1 — 15 out of 620 results