The data conundrum: compression of automotive imaging data and deep neural network based perception

Chan, Pak Hung; Souvalioti, Georgina; Huggett, Anthony; G, Kirsch; Donzella, Valentina

doi:10.2352/issn.2694-118x.2021.lim-78

Cited by 11 publications

(22 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work expands on [2] and evaluates the effect on object detection via compression tuned Faster R-CNN; video compression has been implemented using AVC and HEVC. Additionally, in this work, the use of images compressed using different compression standards for the training and transmitted datasets was investigated for the first time.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Influence of AVC and HEVC compression on detection of vehicles through Faster R-CNN

Chan¹,

Huggett²,

Souvalioti³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

<div><div><div><p>Situational awareness based on the data collected by the vehicle perception sensors (i.e. LiDAR, RADAR, camera and ultrasonic sensors) is key for achieving assisted and automated driving functions in a safe and reliable way. However, the data rates generated by the sensor suite are difficult to support over traditional wired data communication networks on the vehicle, hence there is an interest in techniques that reduce the amount of sensor data to be transmitted without losing key information or introducing unacceptable delays. These techniques must be analysed in combination with the consumer of the data, which will most likely be a machine learning algorithm based on deep neural networks (DNNs). In this paper we demonstrate that by compression tuning the DNNs (i.e. transfer learning by re-training with compressed data) the DNN average precision and recall can significantly improve when uncompressed and compressed data are transmitted. This improvement is achieved independently from the compression standard used for compression-training (we used AVC and HEVC), and also when training and transmitted data use the same compression standard or different compression standards. Furthermore, the performance of the DNNs is stable when transmitting data with increasing lossy compression rate, up to a compression ratio of approximately 1200:1; above this value the performance starts to degrade. This work paves the way for the use of compressed sensor data in assisted and automated driving in combination with the optimisation of compression- tuned DNNs.</p></div></div></div>

show abstract

Section: Methodsmentioning

confidence: 99%

“…object detection and classification, sign recognition, lane centring, etc.) and can leverage mature computer vision and machine learning algorithms to process their data [2][3].…”

Section: Introductionmentioning

confidence: 99%

Influence of AVC and HEVC compression on detection of vehicles through Faster R-CNN

Chan¹,

Huggett²,

Souvalioti³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Semantic segmentation is an important task in computer vision that can be used in many applications, and it is based on assigning to each pixel in a frame a specific class [14]. Most of the existing works evaluate the influence on quality by considering the implications on object detection, without considering semantic segmentation [4], [5]. Therefore, we also evaluate the performance of the semantic segmentation on the compressed data with more than 20 classes and 8 categories.…”

Section: Region-based Video Compressionmentioning

confidence: 99%

“…However, this standard was designed for human vision, and the implications on machine learning (ML) based perception have been only recently started to be discussed and analysed (e.g. retraining deep neural networks with compressed data) [4], [5]. Therefore it is clear that more work needs to be done to exhaustively investigate the relationship between video compression and ML based perception for automated driving.…”

Section: Introductionmentioning

confidence: 99%

A Two-stage H.264 based Video Compression Method for Automotive Cameras

Wang

Chan

Donzella

2022

2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS)

Self Cite

View full text Add to dashboard Cite

“…V5 [71]); the results are not reported here for conciseness, and they show the same trends of displayed performance in sec. V [72]. In all cases, the NNs have been re-trained and optimised using the original dataset; then they have been used to evaluate the original and variant datasets (assuming that pre-trained NNs will be deployed for perception in assisted and automated driving functions).…”

Section: B Nn Based Object Detectormentioning

confidence: 99%

Analysis of Automotive Camera Sensor Noise Factors

Li¹,

Chan²,

Baris³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

<p>Assisted and automated driving functions are increasingly deployed to support improved safety, efficiency, and enhance driver experience. However, there are still key technical challenges that need to be overcome, such as the degradation of perception sensor data due to noise factors. The quality of data being generated by sensors can directly impact the planning and control of the vehicle, which can affect the vehicle safety. This work builds on a recently proposed framework, analysing noise factors on automotive LiDAR sensors, and deploys it to camera sensors, focusing on the specific disturbed sensor outputs via a detailed analysis and classification of automotive camera specific noise sources (30 noise factors are identified and classified in this work). Moreover, the noise factor analysis has recognised two omnipresent and independent noise factors (i.e. obstruction and windshield distortion). These noise factors have been modelled to generate noisy camera data; their impact on the perception step, based on deep neural networks, has been evaluated when the noise factors are applied independently and simultaneously. It is demonstrated that the performance degradation from the combination of noise factors is not simply the accumulated performance degradation from each single factor, which raises the importance of including the simultaneous analysis of multiple noise factors. Thus, the framework can support and enhance the use of simulation for development and testing of automated vehicles through careful consideration of the noise factors affecting camera data. </p>

show abstract

The data conundrum: compression of automotive imaging data and deep neural network based perception

Cited by 11 publications

References 16 publications

Influence of AVC and HEVC compression on detection of vehicles through Faster R-CNN

Influence of AVC and HEVC compression on detection of vehicles through Faster R-CNN

A Two-stage H.264 based Video Compression Method for Automotive Cameras

Analysis of Automotive Camera Sensor Noise Factors

Contact Info

Product

Resources

About