Published on Fri Jan 29 2016

What Can I Do Around Here? Deep Functional Scene Understanding for Cognitive Robots

Chengxi Ye, Yezhou Yang, Cornelia Fermuller, Yiannis Aloimonos

Robots must have a functional understanding of the visual scene to perform tasks. We address the problem of localizing and recognition of functional areas from an arbitrary indoor scene. We also demonstrate that our detection model can generalized onto novel indoor scenes.

0
0
0
Abstract

For robots that have the capability to interact with the physical environment through their end effectors, understanding the surrounding scenes is not merely a task of image classification or object recognition. To perform actual tasks, it is critical for the robot to have a functional understanding of the visual scene. Here, we address the problem of localizing and recognition of functional areas from an arbitrary indoor scene, formulated as a two-stage deep learning based detection pipeline. A new scene functionality testing-bed, which is complied from two publicly available indoor scene datasets, is used for evaluation. Our method is evaluated quantitatively on the new dataset, demonstrating the ability to perform efficient recognition of functional areas from arbitrary indoor scenes. We also demonstrate that our detection model can be generalized onto novel indoor scenes by cross validating it with the images from two different datasets.

Mon Jan 20 2020
Machine Learning
Real-Time Object Detection and Recognition on Low-Compute Humanoid Robots using Deep Learning
In the near future, humanoid robots would share home space and assist us in our daily and routine activities through object manipulations. The proposed algorithm for object detection and localization is an empirical modification of YOLOv3, based on indoor experiments in multiple scenarios.
0
0
0
Sun Aug 01 2021
Computer Vision
Object-to-Scene: Learning to Transfer Object Knowledge to Indoor Scene Recognition
Object-to-Scene (OTS) method extracts object features and learns object relations to recognize indoor scenes. OTS outperforms the state-of-the-art methods by more than 2% on indoor scene recognition.
1
0
1
Thu Sep 28 2017
Artificial Intelligence
Are we done with object recognition? The iCub robot's perspective
We report on the benefits and limitations of current deep learning approaches to object recognition in robot vision scenarios. We consider a natural human-robot interaction to design a data-acquisition protocol for visual object recognition on the iCub humanoid robot.
0
0
0
Fri May 05 2017
Computer Vision
Bridging between Computer and Robot Vision through Data Augmentation: a Case Study on Object Recognition
This paper proposes a simple yet effective data augmentation layer that zooms on the object of interest and simulates the object detection outcome of a robot vision system. The layer, that can be used with any convolutional deep architecture, brings an increase in object recognition performance of up to 7
0
0
0
Mon Sep 18 2017
Computer Vision
Recognizing Objects In-the-wild: Where Do We Stand?
The ability to recognize objects is an essential skill for a robotic system performing in human-populated environments. Despite decades of effort, robots are still missing good visual perceptual systems.
0
0
0
Thu Mar 09 2017
Computer Vision
A Self-supervised Learning System for Object Detection using Physics Simulation and Multi-view Pose Estimation
The focus is on detecting objects placed in tight environments, such as a shelf with multiple objects. The models are placed in physically realistic poses with respect to their environment to generate a labeled synthetic dataset. To further improve object detection, the network self-trains over real images that are labeled
0
0
0