Object recognition in noisy RGB-D data using GNG
نویسندگان
چکیده
منابع مشابه
pattern recognition in maintenance data using methodologies data minitng (cade study isfahan regional power electric company)
فعالیت های نگهداری و تعمیرات اطلاعاتی را تولید می کند که می تواند در تعیین زمان های بیکاری و ارایه یک برنامه زمان بندی شده یا تعیین هشدارهای خرابی به پرسنل نگهداری و تعمیرات کمک کند. وقتی که مقدار داده های تولید شده زیاد باشند، فهم بین متغیرها بسیار مشکل می شوند. این پایان نامه به کاربردی از داده کاوی برای کاوش پایگاه های داده چندبعدی در حوزه نگهداری و تعمیرات، برای پیدا کردن خرابی هایی که موجب...
15 صفحه اولRGB-D Object Recognition Using Deep Convolutional Neural Networks
We address the problem of object recognition from RGB-D images using deep convolutional neural networks (CNNs). We advocate the use of 3D CNNs to fully exploit the 3D spatial information in depth images as well as the use of pretrained 2D CNNs to learn features from RGB-D images. There exists currently no large scale dataset available comprising depth information as compared to those for RGB da...
متن کاملHand Gesture Recognition from RGB-D Data using 2D and 3D Convolutional Neural Networks: a comparative study
Despite considerable enhances in recognizing hand gestures from still images, there are still many challenges in the classification of hand gestures in videos. The latter comes with more challenges, including higher computational complexity and arduous task of representing temporal features. Hand movement dynamics, represented by temporal features, have to be extracted by analyzing the total fr...
متن کاملToward Object-based Place Recognition in Dense RGB-D Maps
Longterm localization and mapping requires the ability to detect when places are being revisited to “close loops” and mitigate odometry drift. The appearance-based approaches solve this problem by using visual descriptors to associate camera imagery. This method has proven remarkably successful, yet performance will always degrade with drastic changes in viewpoint or illumination. In this paper...
متن کاملContextual object category recognition for RGB-D scene labeling
Recent advances in computer vision on the one hand, and imaging technologies on the other hand, have opened up a number of interesting possibilities for robust 3D scene labeling. This paper presents contributions in several directions to improve the state-of-the-art in RGB-D scene labeling. First, we present a novel combination of depth and color features to recognize different object categorie...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Pattern Analysis and Applications
سال: 2016
ISSN: 1433-7541,1433-755X
DOI: 10.1007/s10044-016-0546-y