Filters








15,404 Hits in 3.8 sec

Representation Sharing for Fast Object Detector Search and Beyond [article]

Yujie Zhong, Zelu Deng, Sheng Guo, Matthew R. Scott, Weilin Huang
2020 arXiv   pre-print
To cope with the designed search space, a novel search algorithm termed Representation Sharing (RepShare) is proposed to effectively identify the best combinations of the defined transformations.  ...  FAD consists of a designed search space and an efficient architecture search algorithm. The search space contains a rich set of diverse transformations designed specifically for object detection.  ...  In this work, we develop an efficient NAS algorithm for object detectors, by fast searching the optimized transformations.  ... 
arXiv:2007.12075v4 fatcat:wj2hhhnyrfgivg2wr5sx3msgsm

S-CNN: Subcategory-Aware Convolutional Networks for Object Detection

Tao Chen, Shijian Lu, Jiayuan Fan
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
An iterative learning algorithm is designed for the joint optimization of image subcategorization, multi-component ACF detector, and subcategory-aware CNN classifier.  ...  The marriage between the deep convolutional neural network (CNN) and region proposals has made breakthroughs for object detection in recent years.  ...  , e.g. selective search [15] in R-CNN and Fast R-CNN and region proposal network (RPN) [10] in Faster R-CNN.  ... 
doi:10.1109/tpami.2017.2756936 pmid:28961103 fatcat:puy3zl6uonfnfpjf4ogqyw5i3q

Scale Optimization for Full-Image-CNN Vehicle Detection [article]

Yang Gao, Shouyan Guo, Kaimin Huang, Jiaxin Chen, Qian Gong, Yang Zou, Tong Bai, Gary Overett
2018 arXiv   pre-print
In this paper we present further study of the use and adaptation of the Faster R-CNN object detection method for datasets presenting natural scale distribution and unbiased real-world object frequency.  ...  We significantly increase detection AP for the KITTI dataset car class from 76.3% on our baseline Faster R-CNN detector to 83.6% in our improved detector.  ...  Fast R-CNN uses a traditional Selective Search [15] to generate region proposals. This method is computationally expensive and is a bottleneck for real-time processing.  ... 
arXiv:1802.06926v1 fatcat:ejwgnvygbjhaxg5istt6kj4vzi

Scale-Aware Trident Networks for Object Detection [article]

Yanghao Li, Yuntao Chen, Naiyan Wang, Zhaoxiang Zhang
2019 arXiv   pre-print
As a bonus, a fast approximation version of TridentNet could achieve significant improvements without any additional parameters and computational cost compared with the vanilla detector.  ...  Then, we adopt a scale-aware training scheme to specialize each branch by sampling object instances of proper scales for training.  ...  The RPN and Fast R-CNN heads are shared among branches and ignored for simplicity.  ... 
arXiv:1901.01892v2 fatcat:mb5n2jo5zngoxbspmba4oydnha

Deep Learning for Generic Object Detection: A Survey [article]

Li Liu, Wanli Ouyang, Xiaogang Wang, Paul Fieguth, Jie Chen, Xinwang Liu, Matti Pietikäinen
2019 arXiv   pre-print
Deep learning techniques have emerged as a powerful strategy for learning feature representations directly from data and have led to remarkable breakthroughs in the field of generic object detection.  ...  More than 300 research contributions are included in this survey, covering many aspects of generic object detection: detection frameworks, object feature representation, object proposal generation, context  ...  Acknowledgments The authors would like to thank the pioneer researchers in generic object detection and other related fields.  ... 
arXiv:1809.02165v4 fatcat:b7ozzcy46bek5jx7l3qomj6e3q

DeepProposals: Hunting Objects and Actions by Cascading Deep Convolutional Layers [article]

Amir Ghodrati, Ali Diba, Marco Pedersoli, Tinne Tuytelaars, Luc Van Gool
2016 arXiv   pre-print
In this paper, a new method for generating object and action proposals in images and videos is proposed.  ...  We show that our DeepProposals outperform most of the previously proposed object proposal and action proposal approaches and, when plugged into a CNN-based object detector, produce state-of-the-art detection  ...  Acknowledgements This work was supported by DBOF PhD scholarship, KU Leuven CAMETRON project and FWO project "Monitoring of Abnormal Activity with Camera Systems".  ... 
arXiv:1606.04702v1 fatcat:2sf6zgw4lfg4bggmhroyx6er2y

No Spare Parts: Sharing Part Detectors for Image Categorization [article]

Pascal Mettes, Jan C. van Gemert, Cees G. M. Snoek
2016 arXiv   pre-print
This work aims for image categorization using a representation of distinctive parts.  ...  Part selection should not be done separately for each category, but instead be shared and optimized over all categories.  ...  Acknowledgements This research is supported by the STW STORY project and the Dutch national program COMMIT. References  ... 
arXiv:1510.04908v2 fatcat:omcpr6qpujb33mjaw7wcjwm2ou

Deep Learning for Generic Object Detection: A Survey

Li Liu, Wanli Ouyang, Xiaogang Wang, Paul Fieguth, Jie Chen, Xinwang Liu, Matti Pietikäinen
2019 International Journal of Computer Vision  
Deep learning techniques have emerged as a powerful strategy for learning feature representations directly from data and have led to remarkable breakthroughs in the field of generic object detection.  ...  More than 300 research contributions are included in this survey, covering many aspects of generic object detection: detection frameworks, object feature representation, object proposal generation, context  ...  The authors would like to thank the pioneering researchers in generic object detection and other related fields.  ... 
doi:10.1007/s11263-019-01247-4 fatcat:isdmz4febvbthgowo33c6ifhm4

Playing Doom with SLAM-Augmented Deep Reinforcement Learning [article]

Shehroze Bhatti, Alban Desmaison, Ondrej Miksik, Nantas Nardelli, N. Siddharth, Philip H. S. Torr
2016 arXiv   pre-print
The different components are automatically extracted and composed into a topological representation using on-the-fly object detection and 3D-scene reconstruction.We evaluate the efficacy of our approach  ...  We augment the raw image input to a Deep Q-Learning Network (DQN), by adding details of objects and structural elements encountered, along with the agent's localisation.  ...  Finally the Faster-RCNN [36] combines the region proposals and object detector into a single unified network trainable end-to-end with shared convolutional features which leads to very fast detection  ... 
arXiv:1612.00380v1 fatcat:k5qqrfl5lrdajeesnnnn4mbovq

Machine Learning for the LHCb Simulation [article]

Lucio Anderlini
2022 arXiv   pre-print
In this contribution, I discuss how Machine Learning can help to speed up the Detector Simulation for the upcoming Runs of the LHCb experiment.  ...  reconstruction and selection efficiency.  ...  Nonetheless, the method had success enabling fast simulation of detectors beyond calorimeters, providing for example the output of the LHCb RICH detectors [32] and muon system [33] .  ... 
arXiv:2110.07925v2 fatcat:qmpgxublczdgxnk2kih5wi7o2u

Tattoo Image Search at Scale: Joint Detection and Compact Representation Learning

Hu Han, Jie Li, Anil K. Jain, shiguang shan, Xilin Chen
2019 IEEE Transactions on Pattern Analysis and Machine Intelligence  
While the features in the backbone network are shared by both tattoo detection and compact representation learning, individual latent layers of each sub-network optimize the shared features toward the  ...  The explosive growth of digital images in video surveillance and social media has led to the significant need for efficient search of persons of interest in law enforcement and forensic applications.  ...  ) (grant GJHZ1843), and Youth Innovation Promotion Association CAS (2018135).  ... 
doi:10.1109/tpami.2019.2891584 pmid:30629491 fatcat:zgkocarum5dmfieo4c2si44xli

A Survey of Modern Deep Learning based Object Detection Models [article]

Syed Sahil Abbas Zaidi, Mohammad Samar Ansari, Asra Aslam, Nadia Kanwal, Mamoona Asghar, Brian Lee
2021 arXiv   pre-print
This article surveys recent developments in deep learning based object detectors.  ...  Object Detection is the task of classification and localization of objects in an image or video. It has gained prominence in recent years due to its widespread applications.  ...  a fast and easy to train object detector that could work in existing production systems.  ... 
arXiv:2104.11892v2 fatcat:a3y5rzeqvjf2vbyxm42e76z2tq

R-FCN-3000 at 30fps: Decoupling Detection and Classification

Bharat Singh, Hengduo Li, Abhishek Sharma, Larry S. Davis
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We exploit the fact that many object classes are visually similar and share parts.  ...  Our approach is a modification of the R-FCN architecture to learn shared filters for performing localization across different object classes.  ...  Indeed, the very first application of deep-learning for object detection [13] used Selective-Search [39] to obtain class-agnostic object proposals and classified them using a deep CNN -fine-tuned AlexNet  ... 
doi:10.1109/cvpr.2018.00119 dblp:conf/cvpr/SinghLSD18 fatcat:5xrhlqiy2ncnlgethzgdvfgypy

DeepProposals: Hunting Objects and Actions by Cascading Deep Convolutional Layers

Amir Ghodrati, Ali Diba, Marco Pedersoli, Tinne Tuytelaars, Luc Van Gool
2017 International Journal of Computer Vision  
In this paper, a new method for generating object and action proposals in images and videos is proposed.  ...  We show that DeepProposals outperform most of the previous object proposal and action proposal approaches and, when plugged into a CNN-based object detector, produce state-of-the-art detection performance  ...  reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.  ... 
doi:10.1007/s11263-017-1006-x fatcat:6t7aalbimfh63hy6kcifwy3zwy

A Goal Reasoning Agent for Controlling UAVs in Beyond-Visual-Range Air Combat

Michael W. Floyd, Justin Karneeb, Philip Moore, David W. Aha
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
The TBM incorporates goal reasoning, automated planning, opponent behavior recognition, state prediction, and discrepancy detection to operate in a real-time, dynamic, uncertain, and adversarial environment  ...  We describe the Tactical Battle Manager (TBM), an intelligent agent that uses several integrated artificial intelligence techniques to control an autonomous unmanned aerial vehicle in simulated beyond-visual-range  ...  Acknowledgments Thanks to OSD ASD (R&E) for supporting this research, and our subject matter experts for their many contributions.  ... 
doi:10.24963/ijcai.2017/657 dblp:conf/ijcai/FloydKMA17 fatcat:lzntelzlkfb3vf7uocmf3xe33m
« Previous Showing results 1 — 15 out of 15,404 results