Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance
As difficult vision-based tasks like object detection and monocular depth estimation are making their way in real-time applications and as more light weighted solutions for autonomous vehicles navigation systems are emerging, obstacle detection and collision prediction are two very challenging tasks...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/7/4/61 |
| _version_ | 1797540034266529792 |
|---|---|
| author | David Urban Alice Caplier |
| author_facet | David Urban Alice Caplier |
| author_sort | David Urban |
| collection | DOAJ |
| description | As difficult vision-based tasks like object detection and monocular depth estimation are making their way in real-time applications and as more light weighted solutions for autonomous vehicles navigation systems are emerging, obstacle detection and collision prediction are two very challenging tasks for small embedded devices like drones. We propose a novel light weighted and time-efficient vision-based solution to predict Time-to-Collision from a monocular video camera embedded in a smartglasses device as a module of a navigation system for visually impaired pedestrians. It consists of two modules: a static data extractor made of a convolutional neural network to predict the obstacle position and distance and a dynamic data extractor that stacks the obstacle data from multiple frames and predicts the Time-to-Collision with a simple fully connected neural network. This paper focuses on the Time-to-Collision network’s ability to adapt to new sceneries with different types of obstacles with supervised learning. |
| first_indexed | 2024-03-10T12:54:19Z |
| format | Article |
| id | doaj.art-865aba8a8fd14beeb26c8f4d9456d6cc |
| institution | Directory Open Access Journal |
| issn | 2313-433X |
| language | English |
| last_indexed | 2024-03-10T12:54:19Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj.art-865aba8a8fd14beeb26c8f4d9456d6cc2023-11-21T12:03:16ZengMDPI AGJournal of Imaging2313-433X2021-03-01746110.3390/jimaging7040061Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles AvoidanceDavid Urban0Alice Caplier1CNRS, GIPSA-Lab, Institute of Engineering, University of Grenoble Alpes, 38000 Grenoble, FranceCNRS, GIPSA-Lab, Institute of Engineering, University of Grenoble Alpes, 38000 Grenoble, FranceAs difficult vision-based tasks like object detection and monocular depth estimation are making their way in real-time applications and as more light weighted solutions for autonomous vehicles navigation systems are emerging, obstacle detection and collision prediction are two very challenging tasks for small embedded devices like drones. We propose a novel light weighted and time-efficient vision-based solution to predict Time-to-Collision from a monocular video camera embedded in a smartglasses device as a module of a navigation system for visually impaired pedestrians. It consists of two modules: a static data extractor made of a convolutional neural network to predict the obstacle position and distance and a dynamic data extractor that stacks the obstacle data from multiple frames and predicts the Time-to-Collision with a simple fully connected neural network. This paper focuses on the Time-to-Collision network’s ability to adapt to new sceneries with different types of obstacles with supervised learning.https://www.mdpi.com/2313-433X/7/4/61deep learningcollision detectionTime-to-Collision predictionreal-timeobject detectionmonocular depth estimation |
| spellingShingle | David Urban Alice Caplier Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance Journal of Imaging deep learning collision detection Time-to-Collision prediction real-time object detection monocular depth estimation |
| title | Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance |
| title_full | Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance |
| title_fullStr | Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance |
| title_full_unstemmed | Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance |
| title_short | Time- and Resource-Efficient Time-to-Collision Forecasting for Indoor Pedestrian Obstacles Avoidance |
| title_sort | time and resource efficient time to collision forecasting for indoor pedestrian obstacles avoidance |
| topic | deep learning collision detection Time-to-Collision prediction real-time object detection monocular depth estimation |
| url | https://www.mdpi.com/2313-433X/7/4/61 |
| work_keys_str_mv | AT davidurban timeandresourceefficienttimetocollisionforecastingforindoorpedestrianobstaclesavoidance AT alicecaplier timeandresourceefficienttimetocollisionforecastingforindoorpedestrianobstaclesavoidance |