Filters








17,488 Hits in 4.1 sec

Detecting Humans in RGB-D Data with CNNs [article]

Kaiyang Zhou, Adeline Paiement, Majid Mirmehdi
2022 arXiv   pre-print
We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs.  ...  We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.  ...  of people detection in RGB-D data with CNNs are in demand.  ... 
arXiv:2207.08064v1 fatcat:a25nmxc2xjbxbpq6xwiki63u4i

Detecting humans in RGB-D data with CNNs

Kaiyang Zhou, Adeline Paiement, Majid Mirmehdi
2017 2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA)  
of people detection in RGB-D data with CNNs are in demand.  ...  Introduction RGB-D images encapsulate richer information by providing depth along with color values. In RGB-D human detection, depth information is usually used to reduce the search space [1] .  ... 
doi:10.23919/mva.2017.7986862 dblp:conf/mva/ZhouPM17 fatcat:q7d6ve5q7vcv5kk2bcbn7cb6jq

Action recognition based on a mixture of RGB and depth based skeleton

Srijan Das, Michal Koperski, Francois Bremond, Gianpiero Francesca
2017 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)  
In this paper, we compare two skeleton detection methods: the depthmap based method used with Kinect camera and RGB based method that uses Deep Convolutional Neural Networks.  ...  In order to balance the pros and cons of mentioned skeleton detection methods w.r.t. action recognition task, we propose a fusion of classifiers trained based on each skeleton detection method.  ...  We use χ 2 kernel from these CNN features to classify the actions. We show that both the skeleton detection methods carry complementary information as fusion improves the results.  ... 
doi:10.1109/avss.2017.8078548 dblp:conf/avss/DasKBF17 fatcat:c2piwcmjjvc5beg7joke7ptzsa

Dynamic Obstacle Detection in Traffic Environments

Gopi Krishna Erabati, Helder Araujo
2019 Proceedings of the 13th International Conference on Distributed Smart Cameras - ICDSC 2019  
This paper presents a comparison of state-of-art object detection techniques like Faster R-CNN, YOLO and SSD with 2D image data.  ...  The proposed model incorporates two stage architecture modality for RGB and depth processing and later fused hierarchically. The model will be trained and tested on RGB-D dataset in the future.  ...  Object Detection with 3D data -proposal and future work 3D data (RGB + depth) is being used in many applications with the advent of low cost RGB-D cameras.  ... 
doi:10.1145/3349801.3357134 dblp:conf/icdsc/ErabatiA19 fatcat:d6jqxgovf5gfhlfldmuilvzmdi

Deeply Exploit Depth Information for Object Detection

Saihui Hou, Zilei Wang, Feng Wu
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
This paper addresses the issue on how to more effectively coordinate the depth with RGB aiming at boosting the performance of RGB-D object detection.  ...  Actually, such a detection way is in accordance with the mechanism of the primary neural cortex (V 1 ) in brain.  ...  For the RGB-D object detection with CNN, the key is how to elegantly coordinate the RGB with depth information in feature learning.  ... 
doi:10.1109/cvprw.2016.140 dblp:conf/cvpr/HouWW16 fatcat:44htjmnv3vbylm45gsef2x5vta

Convolutional Neural Network Using Kalman Filter for Human Detection and Tracking on RGB-D Video

Jovin Angelico, Ken Ratri Retno Wardani
2018 CommIT Journal  
For training and testing purposes, there are two kinds of RGB-D datasets used with different points of view and lighting conditions.  ...  The computer ability to detect human being by computer vision is still being improved both in accuracy or computation time. In low-lighting condition, the detection accuracy is usually low.  ...  This research uses RGB-D datasets and combines CNN method with Kalman filter to detect and track humans.  ... 
doi:10.21512/commit.v12i2.4890 fatcat:flm5nwsye5gizprzbvawcd3u7m

Using Deep Learning to Find Victims in Unknown Cluttered Urban Search and Rescue Environments

Angus Fung, Long Yu Wang, Kaicheng Zhang, Goldie Nejat, Beno Benhabib
2020 Current Robotics Reports  
Summary End-to-end deep networks can be used for finding victims in USAR by autonomously extracting RGB-D image features from sensory data.  ...  Experimental results show that RetinaNet has the highest mean average precision (77.66%) and recall (86.98%) for detecting victims with body part occlusions in different lighting conditions.  ...  In general, using RGB-D information resulted in higher precision-recall compared to only using RGB or depth data.  ... 
doi:10.1007/s43154-020-00011-8 fatcat:ywz4n2zvovbirducckjsm2nl2m

Deeply Exploit Depth Information for Object Detection [article]

Saihui Hou, Zilei Wang, Feng Wu
2016 arXiv   pre-print
This paper addresses the issue on how to more effectively coordinate the depth with RGB aiming at boosting the performance of RGB-D object detection.  ...  Actually, such a detection way is in accordance with the mechanism of the primary neural cortex (V1) in brain.  ...  For the RGB-D object detection with CNN, the key is how to elegantly coordinate the RGB with depth information in feature learning.  ... 
arXiv:1605.02260v1 fatcat:rffryvwzdvcbbnrrocqwlkzbsu

RGB-D Data-Based Action Recognition: A Review

Muhammad Bilal Shaikh, Douglas Chai
2021 Sensors  
In this paper, we focus solely on data fusion and recognition techniques in the context of vision with an RGB-D perspective.  ...  Classification of human actions is an ongoing research problem in computer vision.  ...  Education and Learning Classifying human actions from RGB-D data plays an important role in education and learning.  ... 
doi:10.3390/s21124246 fatcat:7dvocdy63rckne5yunhfsnr4p4

Fail-Safe Human Detection for Drones Using a Multi-Modal Curriculum Learning Approach

Ali Safa, Tim Verbelen, Ilja Ocket, Andre Bourdoux, Francky Catthoor, Georges Gielen
2021 IEEE Robotics and Automation Letters  
In addition, we propose a baseline CNN architecture with crossfusion highways and introduce a curriculum learning strategy for multi-modal data termed SAUL, which greatly enhances the robustness of the  ...  In order to enable the fusion of radars with both event-based and standard cameras, we present KUL-UAVSAFE, a first-of-its-kind dataset for the study of safety-critical people detection by drones.  ...  Then, fine-tuning the detection head, in turn, results in a high-accuracy human detection output. D.  ... 
doi:10.1109/lra.2021.3125450 fatcat:z27e6jqrlnaedoezds4exewxey

Detecting object affordances with Convolutional Neural Networks

Anh Nguyen, Dimitrios Kanoulas, Darwin G. Caldwell, Nikos G. Tsagarakis
2016 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
We present a novel and real-time method to detect object affordances from RGB-D images.  ...  Our method trains a deep Convolutional Neural Network (CNN) to learn deep features from the input data in an end-to-end manner.  ...  20] or RGB-D data [21] .  ... 
doi:10.1109/iros.2016.7759429 dblp:conf/iros/NguyenKCT16a fatcat:jvhnlje7rjbidbacihywjtcg3u

Vision and Inertial Sensing Fusion for Human Action Recognition : A Review [article]

Sharmin Majumder, Nasser Kehtarnavaz
2020 arXiv   pre-print
Human action recognition is used in many applications such as video surveillance, human computer interaction, assistive living, and gaming.  ...  This paper provides a survey of the papers in which vision and inertial sensing are used simultaneously within a fusion framework in order to perform human action recognition.  ...  Depth data for the UTD-MHAD dataset were captured by a Kinect RGB-D camera.  ... 
arXiv:2008.00380v1 fatcat:vvjkrkvkpvbctpi63zoax4pcue

Multi-Glimpse LSTM with Color-Depth Feature Fusion for Human Detection [article]

Hengduo Li, Jun Liu, Guyue Zhang, Yuan Gao, Yirui Wu
2017 arXiv   pre-print
With the development of depth cameras such as Kinect and Intel Realsense, RGB-D based human detection receives continuous research attention due to its usage in a variety of applications.  ...  To the best of our knowledge, this is the first attempt to utilize LSTM structure for RGB-D based human detection. Our method achieves superior performance on two publicly available datasets.  ...  [16] also explored to apply neural networks for RGB-D based human detection and tracking. A deep CNN was used in their method to identify generated proposals.  ... 
arXiv:1711.01062v1 fatcat:vajlsm6j2ncwralzpjk3iveeou

Joint Distance Maps Based Action Recognition With Convolutional Neural Networks

Chuankun Li, Yonghong Hou, Pichao Wang, Wanqing Li
2017 IEEE Signal Processing Letters  
With a simple 7-layer network, we obtain 89.3% accuracy on validation set of the NTU RGB+D dataset.  ...  In this paper, we propose a novel convolutional neural networks (CNN) based framework for both action classification and detection.  ...  Faster R-CNN was originally designed for object detection in still images. [9] adapted the framework to temporal activity detection in RGB videos with a 3D convolutional network.  ... 
doi:10.1109/lsp.2017.2678539 fatcat:5uojichfkffrtmhhdan7mosuwi

Deep Neural Network–Based Double-Check Method for Fall Detection Using IMU-L Sensor and RGB Camera Data

Deok-Won Lee, Kooksung Jun, Khawar Naheem, Mun S. Kim
2021 IEEE Access  
In order to overcome these limitations and detect falls with 100% accuracy, a double-check method for fall detection in elderly people via an inertial measurement unit-location (IMU-L) sensor and a red-green-blue  ...  When a potential fall occurs, the individual's location information is synchronized with the motion data.  ...  In each case, the RGB data were obtained along with the IMU sensor data. Fall detection using RGB images is the final step in detecting falls in the proposed method.  ... 
doi:10.1109/access.2021.3065105 fatcat:44zbf5d6lffhbfosyi4oo75ryq
« Previous Showing results 1 — 15 out of 17,488 results