کنفرانس مهندسی برق ایران

صفحه اصلی / سی و دومین کنفرانس بین المللی مهندسی برق

Scene Understanding in Pick-and-Place Tasks: Analyzing Transformations Between Initial and Final Scenes

نویسندگان :

Seraj Ghasemi¹ Hamed Hosseini² MohammadHossein Koosheshi³ Mehdi Tale Masouleh⁴ Ahmad Kalhor⁵

1- University of Tehran 2- دانشگاه تهران 3- دانشگاه تهران 4- University of Tehran 5- University of Tehran

کلمات کلیدی :

Scene Understanding،Convolutional Neural Networks،Object Detection،Pick-and-Place،YOLOv5

چکیده :

With robots increasingly collaborating with humans in everyday tasks, it is important to take steps toward robotic systems capable of understanding the environment. This work focuses on scene understanding to detect pick-and-place tasks given initial and final images from the scene. To that end, a dataset is collected for object detection and pick-and-place task detection. A YOLOv5 network is subsequently trained to detect the objects in the initial and final scenes. Given the detected objects and their bounding boxes, two methods are proposed to detect the pick-and-place tasks that transform the initial scene into the final scene. A geometric method is proposed that tracks objects' movements in the two scenes and works based on the intersection of the bounding boxes that moved within scenes. Contrarily, the CNN-based method utilizes a Convolutional Neural Network to classify objects with intersected bounding boxes into 5 classes, showing the spatial relationship between the involved objects. The performed pick-and-place tasks are then derived from analyzing the experiments with both scenes. Results show the CNN-based method outscores the geometric method by roughly 12\% in certain scenarios, with an overall success rate of 84.3\%.

لیست مقالات

لیست مقالات بایگانی شده

An Iterative Post-processing Method for Speech Source Separation in Realistic Scenarios

Iman Shahriari - Hossein Zeinali

SGG-Net: Skeleton and Graph-Based Neural Network Approaches for Grasping Objects

AliReza Beigy - Farbod Azimmohseni - Ali Sabzejou - Mehdi Tale Masouleh - Ahmad Kalhor

Ultra-wideband RCS Reduction Using Checkerboard Configuration of Bed of Nails

Sadegh Sarjoughian - Mohsen Maddahali - Ahmad Bakhtafrouz

مکان یابی اهداف در محیط مختلط دید مستقیم و غیر مستقیم مبتنی بر اندازه گیری های RSS و TOA با مدل احتمالاتی

محمدرضا شمسیان - فریدون بهنیا

تاثیر روش‌های کاهش سناریو و عدم قطعیت‌های چندمتغیره بر عملکرد هاب انرژی

مهسا نعمتی فر - حسین شریف زاده

طراحی سیستم هوشمند تشخیص سطح مذاب قالب در ماشین ریخته‌گری مداوم

محمد رضا رشیدی - سید محمد تقی المدرسی - سعیده ذبحی

Physiotherapy Algorithms on FUM-Physio Robot

Keyvan Tayaranian Marvian - Amir Hossein Nazari - Seyed Mohammad Tahamipour Zarandi - Mohammad Reza Akbarzadeh totonchi - Zahra Soltani - Alireza Akbarzadeh totonchi

Multi-Agents Gaussian Estimation and Coverage Control Client-Server Architecture

Mohammad َAzizian Shishavan - Mahdi Zeinali - Azam Salari

ساخت و مشخصه یابی حسگر گاز مونوکسیدکربن مبتنی بر هتروساختار p-n نیترید کربن گرافیتی متخلخل-اکسید مس

سمیرا جوانمردی - شیرین نصر اصفهانی - محمد حسین شیخی

Application of Transfer Learning in Optimized Filter- Bank Regularized CSP to Classification of EEG Signals with Small Dataset

M. Moein Esfahani - Hossein Sadati

بیشتر