Multi-Modal Sensor Registration for Vehicle Perception Via Deep Neural Networks. Giering, M., Venugopalan, V., & Reddy, K. In IEEE High Performance Extreme Computing Conference (HPEC), pages 1-6, September, 2015.
abstract   bibtex   
The ability to simultaneously leverage multiple modes of sensor information is critical for perception of an automated vehicle's physical surroundings. Spatio-temporal alignment of registration of the incoming information is often a prerequisite to analyzing the fused data. The persistence and reliability of multi-modal registration is therefore the key to the stability of decision support systems ingesting the fused information. LiDAR-video systems like on those many driverless cars are a common example of where keeping the LiDAR and video channels registered to common physical features is important. We develop a deep learning method that takes multiple channels of heterogeneous data, to detect the misalignment of the LiDAR-video inputs. A number of variations were tested on the Ford LiDAR-video driving test data set and will be discussed. To the best of our knowledge the use of multi-modal deep convolutional neural networks for dynamic real-time LiDAR-video registration has not been presented.
@inproceedings{Giering2015Multi-moda,
	abstract = {The ability to simultaneously leverage multiple modes of sensor information is critical for perception of an automated vehicle's physical surroundings. Spatio-temporal alignment of registration of the incoming information is often a prerequisite to analyzing the fused data. The persistence and reliability of multi-modal registration is therefore the key to the stability of decision support systems ingesting the fused information. LiDAR-video systems like on those many driverless cars are a common example of where keeping the LiDAR and video channels registered to common physical features is important. We develop a deep learning method that takes multiple channels of heterogeneous data, to detect the misalignment of the LiDAR-video inputs. A number of variations were tested on the Ford LiDAR-video driving test data set and will be discussed. To the best of our knowledge the use of multi-modal deep convolutional neural networks for dynamic real-time LiDAR-video registration has not been presented.},
	author = {Giering, Michael and Venugopalan, Vivek and Reddy, Kishore},
	booktitle = {IEEE High Performance Extreme Computing Conference (HPEC)},
	date-added = {2020-01-20 18:26:09 -0500},
	date-modified = {2020-01-20 18:26:09 -0500},
	keywords = {image fusion;image registration;neural nets;optical radar;radar imaging;video signal processing;Ford LiDAR-video driving test data set;LiDAR-video input misalignment detection;LiDAR-video systems;automated vehicle physical surroundings;decision support systems;deep-learning method;driverless cars;dynamic real-time LiDAR-video registration;fused data analysis;heterogeneous data;multimodal deep-convolutional neural networks;multimodal registration persistence;multimodal registration reliability;multimodal sensor registration;multiple sensor information modes;physical features;spatio-temporal alignment;vehicle perception;Accuracy;Laser radar;Optical imaging;Optical sensors;Testing;Three-dimensional displays;Training},
	month = sep,
	pages = {1-6},
	title = {{Multi-Modal Sensor Registration for Vehicle Perception Via Deep Neural Networks}},
	year = {2015},
	Bdsk-File-1 = {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},
	Bdsk-Url-1 = {http://dx.doi.org/10.1109/HPEC.2015.7322485}}

Downloads: 0