ODDS: Real-Time Object Detection Using Depth Sensors on Embedded GPUs
Detecting objects that are carried when someone enters or exits a room is very useful for a wide range of smart building applications including safety, security, and energy efficiency. While there has been a significant amount of work on object recognition using large-scale RGB image datasets, RGB cameras are too privacy invasive in many smart building applications and they work poorly in the dark. Additionally, deep object detection networks require powerful and expensive GPUs. We propose a novel system that we call ODDS (Object Detector using a Depth Sensor) that can detect objects in real-time using only raw depth data on an embedded GPU, e.g., NVIDIA Jetson TX1. Hence, our solution is significantly less privacy invasive (even if the sensor is compromised) and less expensive, while maintaining a comparable accuracy with state of the art solutions. Specifically, we resort to training a deep convolutional neural network using raw depth images, with curriculum based learning to improve accuracy by considering the complexity and imbalance in object classes and developing a sparse coding based technique that speeds up the system ~2× with minimal loss of accuracy. Based on a complete implementation and real-world evaluation, we see ODDS achieve 80.14% mean average precision in object detection in real-time (5-6 FPS) on a Jetson TX1.
- Research Organization:
- Robert Bosch LLC, Farmington Hills, MI (United States)
- Sponsoring Organization:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Building Technologies Office
- DOE Contract Number:
- EE0007682
- OSTI ID:
- 1811681
- Journal Information:
- 2018 17th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), Conference: ACM/IEEE Conference on Information Processing in Sensor Networks (IPSN) , Porto, Portugal, April 11-13, 2018.
- Country of Publication:
- United States
- Language:
- English
Indoor Person Identification through Footstep Induced Structural Vibration
|
conference | January 2015 |
FORK: fine grained occupancy estimatoR using kinect on ARM embedded platforms
|
conference | November 2017 |
Fast ConvNets Using Group-Wise Brain Damage
|
conference | June 2016 |
Real-Time Fine Grained Occupancy Estimation Using Depth Sensors on ARM Embedded Platforms
|
conference | April 2017 |
Generating Diverse Image Datasets with Limited Labeling
|
conference | October 2016 |
ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression
|
conference | October 2017 |
Sonicdoor
|
conference | November 2017 |
3D through-wall imaging with unmanned aerial vehicles using wifi
|
conference | April 2017 |
Cross Modal Distillation for Supervision Transfer
|
conference | June 2016 |
Channel Pruning for Accelerating Very Deep Neural Networks
|
conference | October 2017 |
Caffe: Convolutional Architecture for Fast Feature Embedding
|
conference | January 2014 |
Forma Track
|
journal | September 2017 |
Going deeper with convolutions
|
conference | June 2015 |
Discriminative Multi-modal Feature Fusion for RGBD Indoor Scene Recognition
|
conference | June 2016 |
DeepIoT
|
conference | November 2017 |
Scpl
|
conference | April 2013 |
Unsupervised Learning of Visual Representations Using Videos
|
conference | December 2015 |
Curriculum learning
|
conference | January 2009 |
Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers | book | January 2011 |
Towards Scalable Summarization of Consumer Videos Via Sparse Dictionary Selection
|
journal | February 2012 |
Histograms of Oriented Gradients for Human Detection
|
conference | January 2005 |
See all by looking at a few: Sparse modeling for finding representative objects
|
conference | June 2012 |
Indoor Scene Understanding with RGB-D Images: Bottom-up Segmentation, Object Detection and Semantic Segmentation
|
journal | November 2014 |
Visual tracking with online Multiple Instance Learning
|
conference | June 2009 |
Low-Power Radio-Optical Beacons for In-View Recognition
|
conference | September 2015 |
Second-order constrained parametric proposals and sequential search-based structured prediction for semantic segmentation in RGB-D images
|
conference | June 2015 |
SUN RGB-D: A RGB-D scene understanding benchmark suite
|
conference | June 2015 |
You Only Look Once: Unified, Real-Time Object Detection
|
conference | June 2016 |
Curriculum learning of multiple tasks
|
conference | June 2015 |
Similar Records
Real-Time Fine Grained Occupancy Estimation Using Depth Sensors on ARM Embedded Platforms
Integrate Light-Weight Deep Learning Tools with Internet of Things