A lightweight CNN called ConvBEERS restores satellite images with 6.9 dB PSNR gain, boosts downstream object detection by up to 5.1% mAP, and deploys on Xilinx Versal FPGA with 41x lower latency than physics-based pipelines.
Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising.IEEE Transactions on Image Processing, 26(7):3142–3155, 2017
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Rethinking Satellite Image Restoration for Onboard AI: A Lightweight Learning-Based Approach
A lightweight CNN called ConvBEERS restores satellite images with 6.9 dB PSNR gain, boosts downstream object detection by up to 5.1% mAP, and deploys on Xilinx Versal FPGA with 41x lower latency than physics-based pipelines.