Filters








12,570 Hits in 4.3 sec

SCALE-AWARENESS FOR MORE ACCURATE OBJECT DETECTION USING MODIFIED SINGLE SHOT DETECTORS

V. Tsironis, C. Stentoumis, N. Lekkas, A. Nikopoulos
2021 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
more elaborate algorithms, especially if we measure detection performance slightly disregarding box localization.  ...  Our contributions are the adjustments to the detector architecture and the introduction of a novel batch sampling method.  ...  to detect small objects in a context-aware manner .  ... 
doi:10.5194/isprs-archives-xliii-b2-2021-801-2021 fatcat:qk7qpcqjlzhw5ekqgqhwsmuy3e

A semi-supervised learning approach for robust indoor-outdoor detection with smartphones

Valentin Radu, Panagiota Katsikouli, Rik Sarkar, Mahesh K. Marina
2014 Proceedings of the 12th ACM Conference on Embedded Network Sensor Systems - SenSys '14  
The environmental context of a mobile device determines how it is used and how the device can optimize operations for greater efficiency and usability.  ...  We find that a particular semi-supervised learning method called co-training, when suitably engineered, is most effective.  ...  We thank Charles Sutton for insightful discussions on semi-supervised learning; we thank Lama Nachman and anonymous reviewers for many helpful suggestions on improving the paper.  ... 
doi:10.1145/2668332.2668347 dblp:conf/sensys/RaduKSM14 fatcat:dfv3yx54onhclcpd7ebggmuevy

3DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detection [article]

He Wang, Yezhen Cong, Or Litany, Yue Gao, Leonidas J. Guibas
2021 arXiv   pre-print
We therefore propose to use the estimated 3D IoU as a localization metric and set category-aware self-adjusted thresholds to filter poorly localized proposals.  ...  We adopt VoteNet as our backbone detector on indoor datasets while we use PV-RCNN on the autonomous driving dataset, KITTI.  ...  Then the local coordinates of these grid points along with their features are pushed through a PointNet to regress class-aware 3D IoU.  ... 
arXiv:2012.04355v3 fatcat:oqlthqnsbvadzgnqievenubrke

Online updating of context-aware landmark detectors for prostate localization in daily treatment CT images

Xiubin Dai, Yaozong Gao, Dinggang Shen
2015 Medical Physics (Lancaster)  
Methods: To localize the prostate in the daily treatment images, the authors first automatically detect six anatomical landmarks on the prostate boundary by adopting a context-aware landmark detection  ...  Specifically, in this method, a two-layer regression forest is trained as a detector for each target landmark.  ...  The training of the context-aware landmark detectors In the context-aware landmark detection method, we train a two-layer regression forest as the detector for each target landmark.  ... 
doi:10.1118/1.4918755 pmid:25979051 pmcid:PMC4409630 fatcat:zqo6vhzn7rghdeht6zmwsr4czm

Advancing Bag-of-Visual-Words Representations for Lesion Classification in Retinal Images

Ramon Pires, Herbert F. Jelinek, Jacques Wainer, Eduardo Valle, Anderson Rocha, Zsolt Ablonczy
2014 PLoS ONE  
Our BoVW scheme is based on sparse lowlevel feature detection with a Speeded-Up Robust Features (SURF) local descriptor, and mid-level features based on semisoft coding with max pooling.  ...  To assess the accuracy for detecting cases that require referral within one year, the sparse extraction technique associated with semi-soft coding and max pooling obtained an AUC of 94.2+2.0%, outperforming  ...  As we show in the experiments, with ANOVA, the semi-soft coding associated with sparse feature extraction provides a good balance for designing an efficient and effective DR-related lesion detector.  ... 
doi:10.1371/journal.pone.0096814 pmid:24886780 pmcid:PMC4041723 fatcat:ghephfq255f2vkvz7bw4uomaui

Real-Time and Robust 3D Object Detection Within Road-Side LiDARs Using Domain Adaptation [article]

Walter Zimmer, Marcus Grabler, Alois Knoll
2022 arXiv   pre-print
We do several sets of experiments for each module in the DASE-ProPillars detector that show that our model outperforms the SE-ProPillars baseline on the real A9 test set and a semi-synthetic A9 test set  ...  We apply domain adaptation from the semi-synthetic A9-Dataset to the semi-synthetic dataset from the Regensburg Next project by applying transfer learning and achieve a 3D mAP@0.25 of 93.49% on the Car  ...  As conventional 3D convolutional backbones in voxelbased 3D detectors are not able to efficiently capture extensive context information, Voxel Transfomer (VoTr) [22] is proposed to resolve this issue  ... 
arXiv:2204.00132v1 fatcat:w7fxwi4vzng7dcjm6t4d63tn2m

PyramidBox: A Context-assisted Single Shot Face Detector [article]

Xu Tang, Daniel K. Du, Zeqiang He, Jingtuo Liu
2018 arXiv   pre-print
First, we design a novel context anchor to supervise high-level contextual feature learning by a semi-supervised method, which we call it PyramidAnchors.  ...  This paper proposes a novel context-assisted single shot face detector, named PyramidBox to handle the hard face detection problem.  ...  We designed a novel context anchor, named PyramidAnchor, to supervise face detector to learn features from contextual parts around faces.  ... 
arXiv:1803.07737v2 fatcat:qnfqkngrlbenzjpmrptbrwwwey

Introducing Context Awareness in Multi-Target Tracking using Re-Identification Methodologies

V. Lovatsis, A. Dimou, P. Daras
2013 5th International Conference on Imaging for Crime Detection and Prevention (ICDP 2013)  
In this paper, re-identification techniques are exploited to add context awareness to a multi-target tracker and enhance its tracking performance, in an online manner.  ...  A baseline tracker, constructed combining a detector and a classifier, is enhanced with the proposed context-aware labeling.  ...  Re-ID approaches consider subjects as sets of local and global features extracted from a set of images [15] .  ... 
doi:10.1049/ic.2013.0262 dblp:conf/icdp/LovatsisDD13 fatcat:xnu4fkrsibdebc7rj2iblle6by

PyramidBox: A Context-Assisted Single Shot Face Detector [chapter]

Xu Tang, Daniel K. Du, Zeqiang He, Jingtuo Liu
2018 Lecture Notes in Computer Science  
First, we design a novel context anchor to supervise high-level contextual feature learning by a semi-supervised method, which we call it PyramidAnchors.  ...  Second, we propose the Low-level Feature Pyramid Network to combine adequate high-level context semantic feature and Low-level facial feature together, which also allows the PyramidBox to predict faces  ...  We designed a novel context anchor, named PyramidAnchor, to supervise face detector to learn features from contextual parts around faces.  ... 
doi:10.1007/978-3-030-01240-3_49 fatcat:j2j3aztppvbitfysopwsbo65du

A novel forensic image analysis tool for discovering double JPEG compression clues

Ali Taimori, Farbod Razzazi, Alireza Behrad, Ali Ahmadi, Massoud Babaie-Zadeh
2016 Multimedia tools and applications  
Existing detectors only operate in a scenario that the image under investigation is explicitly available in JPEG format.  ...  Our tool discriminates single compressed images from double counterparts, estimates the first quantization in double compression, and localizes tampered regions in a forgery examination.  ...  semi-aware TD-and BU-DJCD, we used the optimal feature dimensions in the latent low-dimensional space equal to d TD r 200 and d BU r 60, respectively.  ... 
doi:10.1007/s11042-016-3409-z fatcat:7kycislweban5hu4rvclhouyiq

Multimodal Diaries

Fernando De la Torre, Carlos Agell
2007 Multimedia and Expo, 2007 IEEE International Conference on  
Two main novelties are proposed: • A system that combines both physical and contextual awareness hardware and software.  ...  . • A semi-supervised temporal clustering (SSTC) algorithm that accurately and efficiently groups large amounts of multimodal data into different activities.  ...  Context Awareness Sensors For context awareness, we have used the Logiteck Quickcam camera to record audio and video, a computer monitoring software to record the programs that the user is running, and  ... 
doi:10.1109/icme.2007.4284781 dblp:conf/icmcs/TorreA07 fatcat:qskydoskhjf5ddmcgki7bw53r4

PyramidBox++: High Performance Detector for Finding Tiny Face [article]

Zhihang Li, Xu Tang, Junyu Han, Jingtuo Liu, Ran He
2019 arXiv   pre-print
A large amount of methods have been put forward where PyramidBox designs an effective data augmentation strategy (Data-anchor-sampling) and context-based module for face detector.  ...  Dual-PyramidAnchors facilitate feature learning by introducing progressive anchor loss.  ...  Benefit from the development of image classification and object detection, we further improve Pyramidbox [20] including data augmentation, feature learning, context-aware prediction module and multi-task  ... 
arXiv:1904.00386v1 fatcat:k7tkzef4knctbdu3x2bjy53uma

Minimum Efforts to Build an End-to-End Spatial-Temporal Action Detector [article]

Lin Sui, Chen-Lin Zhang, Lixin Gu, Feng Han
2022 arXiv   pre-print
Current spatial-temporal action detection methods will first use an object detector to obtain person candidate proposals.  ...  In this paper, we explore the strategy to build an end-to-end spatial-temporal action detector with minimal modifications.  ...  In the actor localization part, we build the feature pyramid on top of keyframes feature from Res3 and Res4 layers.  ... 
arXiv:2206.03064v1 fatcat:l5chwbsr6nb45oeuofrixhz72u

Hierarchical Open-Set Object Detection in Unseen Data

Yeong Hyeon Kim, Dong Kyun Shin, Minhaz Uddin Ahmed, Phill Kyu Rhee
2019 Symmetry  
The hierarchical feature model (HFM) was used to learn a new object class by using collaborative sampling (CS), and open-set-aware active semi-supervised learning (ASSL) algorithms.  ...  We were motivated by the observation that deep features extracted from visual objects show a strong hierarchical clustering property.  ...  Our method combines the incremental open-set aware active semi-supervised learning (ASSL) [16] and the dynamic hierarchical feature model (HFM) update algorithm for effectively grouping unseen objects  ... 
doi:10.3390/sym11101271 fatcat:ml2w2mloqrf4flo6nb76rxqkte

On-Device Content Moderation [article]

Anchal Pandey, Sukumar Moharana, Debi Prasanna Mohanty, Archit Panwar, Dewang Agarwal, Siva Prasad Thota
2021 arXiv   pre-print
We have created anensemble of object detector and classifier for filtering of nudeand semi-nude contents.  ...  categories,namely nude, semi-nude and safe images.  ...  Thus, we use SSD with MobileNetV3 as feature extractor for a lightweight detector. B.  ... 
arXiv:2107.11845v1 fatcat:lztmqlxftva2lmdj5leusuvmke
« Previous Showing results 1 — 15 out of 12,570 results