Filters








299,048 Hits in 6.7 sec

Improving a deep learning based RGB-D object recognition model by ensemble learning

Andreas Aakerberg, Kamal Nasrollahi, Thomas Heder
2017 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA)  
Augmenting RGB images with depth information is a well-known method to significantly improve the recognition accuracy of object recognition models.  ...  However, this method has not been widely explored in combination with deep convolutional neural network based RGB-D object recognition models.  ...  CONCLUSIONS In this work, we have shown that forming an ensemble by combining the softmax probabilities of different complementary CNN based RGB-D object recognition models, with weighted averaging to  ... 
doi:10.1109/ipta.2017.8310101 dblp:conf/ipta/AakerbergNH17 fatcat:xqttlwzpozgj3iykwj252s4tua

Improved machine learning for image category recognition by local color constancy

Hamid Reza Vaezi Joze, Mark S. Drew
2010 2010 IEEE International Conference on Image Processing  
Systems for object detection or recognition in images use machine learning based on image descriptors to distinguish object and scene categories.  ...  In a novel contribution, we ascertain that a combination of local color constancy methods and color invariant descriptors improve the performance of object recognition by as much as more than 10 percent  ...  Many systems for object detection or recognition in images use machine learning based on image descriptors to distinguish object and scene categories.  ... 
doi:10.1109/icip.2010.5651069 dblp:conf/icip/JozeD10 fatcat:3ynan32p7rah7o5maw445u477e

Background Filtering for Improving of Object Detection in Images

Ge Qin, Bogdan Vrusias, Lee Gillam
2010 2010 20th International Conference on Pattern Recognition  
We propose a method for improving object recognition in street scene images by identifying and filtering out background aspects.  ...  Our method is independent of the recognition algorithms used for individual objects, and can be extended to generic object recognition in other environments by adapting other object models.  ...  filtering, recognition results (W/O BKGD) improved by over 40% (dF) for all methods, shown inTable 2.  ... 
doi:10.1109/icpr.2010.231 dblp:conf/icpr/QinVG10 fatcat:cgltxca4irdhhmr6lplcbxtasu

RGB-D and corrupted images in assistive blind systems in smart cities

Amany Yehia, Shereen A. Taie
2022 Bulletin of Electrical Engineering and Informatics  
The first phase introduces a new prepossessing model for automatically handling RGB-D on YOLOv3 with an accuracy of 61.50% in detection and 57.02% in recognition.  ...  This paper introduces two novel ways in object detection that improves YOLOv3 technique to deal with corrupted images and RGB-D images.  ...  Table 1 . 1 Table 1 illustrate the average results of combining RGB and Depth images (RGB-D) were 61.50% in detection and 57.02% in recognition.  ... 
doi:10.11591/eei.v11i4.3770 fatcat:yvivdowmtrb67esfb2vwhg5qby

Object Recognition by Integrating Multiple Image Segmentations [chapter]

Caroline Pantofaru, Cordelia Schmid, Martial Hebert
2008 Lecture Notes in Computer Science  
By integrating the image partition hypotheses in an intuitive combined top-down and bottom-up recognition approach, we improve object and feature support.  ...  This instability has led to advocacy for using multiple segmentations of an image.  ...  Extensions We have shown that a straightforward method for combining multiple segmentations can lead to robust and accurate object recognition and object segmentation.  ... 
doi:10.1007/978-3-540-88690-7_36 fatcat:zf5otqchfjghjl7latpg74pd3i

Classification of fused face images using multilayer perceptron neural network [article]

Debotosh Bhattacharjee, Mrinal Kanti Bhowmik, Mita Nasipuri, Dipak Kumar Basu, Mahantapas Kundu
2010 arXiv   pre-print
This paper presents a concept of image pixel fusion of visual and thermal faces, which can significantly improve the overall performance of a face recognition system.  ...  In the experiments we have used Object Tracking and Classification Beyond Visible Spectrum (OTCBVS) database benchmark thermal and visual face images.  ...  The efficiency of our scheme has been demonstrated on Object Tracking and Classification Beyond Visible spectrum (OTCBVS) benchmark database and recognition rate obtained is 95.07%. V.  ... 
arXiv:1007.0633v1 fatcat:proweasho5au7hbnhaol4huq3e

Gait Recognition based on Inverse Fast Fourier Transform Gaussian and Enhancement Histogram Oriented of Gradient

Siti Zaharah A. Rahman, Siti Norul Huda Sheikh Abdullah, Khairul Akram Zainol Ariffin
2018 International Journal on Advanced Science, Engineering and Information Technology  
Thus, this research proposed a combination of frequency and spatial features based on Inverse Fast Fourier Transform and Histogram of Oriented Gradient (IFFTG-HoG) for gait recognition.  ...  It performed better in gait individual classification as the value of average accuracy for the standard dataset 96.7%, 93.1% and 99.6%compared to HoG method by 94.1%, 85.9% and 96.2% in order.  ...  ACKNOWLEDGMENT The author would like to present his gratitude towards the Faculty of Information Science and Technology, the National University of Malaysia for the help and support that had been accorded  ... 
doi:10.18517/ijaseit.8.4-2.6795 fatcat:4xhvvbanrzhxhp2cmnyefgdxam

Refining object recognition using scene specific object appearance frequencies

Arthur Daniel Costea, Robert Varga, Tiberiu Marita, Sergiu Nedevschi
2011 2011 IEEE 7th International Conference on Intelligent Computer Communication and Processing  
This paper presents an automatic annotation method for multimedia data. Different object and scene recognition methods are analyzed from the literature.  ...  Image processing; automatic annotation; scene recognition; object recognition I.  ...  This paper proposes a method for combining information obtained from scene recognition and object recognition.  ... 
doi:10.1109/iccp.2011.6047866 dblp:conf/iccp2/CosteaVMN11 fatcat:h3nmwwsgnvcd3f6uiadirnkrgi

Cultural event recognition by subregion classification with convolutional neural network

Sungheon Park, Nojun Kwak
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Compared to the state-of-the-art methods for classifying Chalearn Looking at People cultural event recognition database, the proposed methods shows competitive results.  ...  The final classification of a scene is performed by combining the classification results of each extracted region of the scene probabilistically.  ...  As in [20] , possible object locations are extracted via selective search which combines an exhaustive search and segmentation.  ... 
doi:10.1109/cvprw.2015.7301335 dblp:conf/cvpr/ParkK15 fatcat:hj44vyvh3jdbpmoxjpoggejyay

Combination of Salient Object Detection and Image Matching for Object Instance Recognition

Evan Kristia Wigati, Gede Putra Kusuma, Yesun Utomo
2020 Advances in Science, Technology and Engineering Systems  
In this paper, we proposed a combination method between Salient Object Detection and Object Instance Recognition using Image Matching and Geometric Verification.  ...  While the results of the combination method show 1.92% performance improvement than the previous method which is BSIS without Salient Object Detection.  ...  This paper focuses on proposing a method for Object Instance Recognition that combines Salient Object Detection and Image Matching with Geometric Verification.  ... 
doi:10.25046/aj050273 fatcat:fpvi2zhvefdtrdkyxtfghu7ji4

Ensembles of Deep Neural Networks for Action Recognition in Still Images [article]

Sina Mohammadi, Sina Ghofrani Majelan, Shahriar B. Shokouhi
2020 arXiv   pre-print
Despite the fact that notable improvements have been made recently in the field of feature extraction and classification, human action recognition is still challenging, especially in images, in which,  ...  Thus, the methods proposed for recognizing human actions in videos cannot be applied to still images.  ...  [10] classified human actions by adopting a BoVW and an SVM classifier. Some methods modeled human-object interactions for recognizing human activities.  ... 
arXiv:2003.09893v1 fatcat:cg3iyzekqfeb7ay5apu5agyl3e

Action Recognition Network Using Stacked Short-Term Deep Features and Bidirectional Moving Average

Jinsol Ha, Joongchol Shin, Hasil Park, Joonki Paik
2021 Applied Sciences  
In this paper, we create short-term pixel-difference images from the input video, and take the difference images as an input to a bidirectional exponential moving average sub-network to analyze the action  ...  maps using a bidirectional, exponential, moving average sub-network, and (iv) late fusion with a three-dimensional convolutional (C3D) sub-network to improve the accuracy of action recognition.  ...  Data Availability Statement: The UCF101 datasets presented in this study are openly available in [13] and can be found here https://www.crcv.ucf.edu/data/UCF101.php.  ... 
doi:10.3390/app11125563 fatcat:xh6376srl5bo5ccdwzlyomuqoi

Bayesian Place Recognition Based on Bag of Objects for Intelligent Vehicle Localization

xianglong wang, Zhaozheng Hu, Mengchao Mu, Gang Huang, Qianwen Tao
2019 IET Intelligent Transport Systems  
The method achieved 0.37 frame errors, taking an average of 4.7 ms for place recognition. Experimental results demonstrated that the proposed method outperforms the classic BoW and binary BoW.  ...  Place recognition matches a corresponding image from a pre-built image sequence, and it plays an important role in visual map-based vehicle localisation.  ...  The BoO feature is a global image feature that needs to be improved. The semantic information is an important feature for place recognition.  ... 
doi:10.1049/iet-its.2018.5431 fatcat:l62diq5vmba7hkzfgvkzwhfug4

Early recognition of tomato gray leaf spot disease based on MobileNetv2-YOLOv3 model

Jun Liu, Xuewei Wang
2020 Plant Methods  
The pre-training method combining mixup training and transfer learning is used to improve the generalisation ability of the model.  ...  This study proposes an early recognition method of tomato leaf spot based on MobileNetv2-YOLOv3 model to achieve a good balance between the accuracy and real-time detection of tomato gray leaf spot.  ...  Acknowledgements Appreciations are given to the editors and reviewer of the Journal Plant Method.  ... 
doi:10.1186/s13007-020-00624-2 pmid:32523613 pmcid:PMC7281931 fatcat:u6isu63wrjcxtbf5rdmnrxqgaq

Enhance Egocentric Grasp Recognition Based Flex Sensor Under Low Illumination

Chana Chansri, Jakkree Srinonchat
2022 Computers Materials & Continua  
The average voltages are used to configure the weighting parameter to improve images in the image enhancement stage.  ...  Initially, a flex sensor is installed to the thumb for object manipulation.  ...  and expertise.  ... 
doi:10.32604/cmc.2022.024026 fatcat:ql4o24crwjfwrifznegm7zkb5m
« Previous Showing results 1 — 15 out of 299,048 results