High Efficiency Compression for Object Detection

Choi, Hyomin; Bajić, Ivan V.

doi:10.1109/icassp.2018.8462653

Cited by 47 publications

(20 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature compression can further improve the efficiency of collaborative intelligence by minimizing the latency and energy of feature data transfer. The impact of compressing the input has been studied in several CNN applications [5,6,7] and the effects vary from case to case. However, the impact of feature compression has not been studied yet, to our knowledge.…”

Section: Introductionmentioning

confidence: 99%

Deep Feature Compression for Collaborative Object Detection

Choi

Bajić

2018

2018 25th IEEE International Conference on Image Processing (ICIP)

Self Cite

126

103

View full text Add to dashboard Cite

Recent studies have shown that the efficiency of deep neural networks in mobile applications can be significantly improved by distributing the computational workload between the mobile device and the cloud. This paradigm, termed collaborative intelligence, involves communicating feature data between the mobile and the cloud. The efficiency of such approach can be further improved by lossy compression of feature data, which has not been examined to date. In this work we focus on collaborative object detection and study the impact of both near-lossless and lossy compression of feature data on its accuracy. We also propose a strategy for improving the accuracy under lossy feature compression. Experiments indicate that using this strategy, the communication overhead can be reduced by up to 70% without sacrificing accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Feature Compression for Collaborative Object Detection

Choi

Bajić

2018

2018 25th IEEE International Conference on Image Processing (ICIP)

Self Cite

126

103

View full text Add to dashboard Cite

show abstract

“…To tackle this problem, MPEG founded an ad-hoc group in 2019 [6]. Promising approaches to increase the coding performance for such applications are, e.g., based on saliency coding as in [7] and [8]. Another approach in [9], optimizes the in-loop filtering of VVC for the VCM task.…”

Section: Decision Algorithmmentioning

confidence: 99%

Robust Deep Neural Object Detection and Segmentation for Automotive Driving Scenario with Compressed Image Data

Fischer

Blum

Herglotz

et al. 2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

Deep neural object detection or segmentation networks are commonly trained with pristine, uncompressed data. However, in practical applications the input images are usually deteriorated by compression that is applied to efficiently transmit the data. Thus, we propose to add deteriorated images to the training process in order to increase the robustness of the two state-of-the-art networks Faster and Mask R-CNN. Throughout our paper, we investigate an autonomous driving scenario by evaluating the newly trained models on the Cityscapes dataset that has been compressed with the upcoming video coding standard Versatile Video Coding (VVC). When employing the models that have been trained with the proposed method, the weighted average precision of the R-CNNs can be increased by up to 3.68 percentage points for compressed input images, which corresponds to bitrate savings of nearly 48 %.

show abstract

“…For less critical automatic video processing applications, e.g., the remote vehicle control in emergency situations a lossy compression of the registered data is possible [ 5 , 6 ]. However, such data loss can substantially affect accuracy of the automatic video processing algorithms [ 18 , 19 , 20 ].…”

Section: Lossless Compression For Adas and Adsmentioning

confidence: 99%

Highly Efficient Lossless Coding for High Dynamic Range Red, Clear, Clear, Clear Image Sensors

Pawlowski

Piniarski

Dąbrowski

2021

Sensors

View full text Add to dashboard Cite

In this paper we present a highly efficient coding procedure, specially designed and dedicated to operate with high dynamic range (HDR) RCCC (red, clear, clear, clear) image sensors used mainly in advanced driver-assistance systems (ADAS) and autonomous driving systems (ADS). The coding procedure can be used for a lossless reduction of data volume under developing and testing of video processing algorithms, e.g., in software in-the-loop (SiL) or hardware in-the-loop (HiL) conditions. Therefore, it was designed to achieve both: the state-of-the-art compression ratios and real-time compression feasibility. In tests we utilized FFV1 lossless codec and proved efficiency of up to 81 fps (frames per second) for compression and 87 fps for decompression performed on a single Intel i7 CPU.

show abstract

High Efficiency Compression for Object Detection

Cited by 47 publications

References 12 publications

Deep Feature Compression for Collaborative Object Detection

Deep Feature Compression for Collaborative Object Detection

Robust Deep Neural Object Detection and Segmentation for Automotive Driving Scenario with Compressed Image Data

Highly Efficient Lossless Coding for High Dynamic Range Red, Clear, Clear, Clear Image Sensors

Contact Info

Product

Resources

About