Atrous Space Bender U-Net (ASBU-Net/LogiNet)
This work addresses the need for fast and efficient semantic segmentation models for deployment on embedded hardware, but it appears incremental as it builds on existing CNN advancements without introducing new paradigm shifts.
The authors tackled the problem of semantic segmentation in high-resolution images under challenging conditions like crowded scenes and occlusion, introducing ASBU-Net, which achieves strong performance with efficient computation and memory usage, as demonstrated in experiments on resource-accuracy trade-offs.
$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.