LCA-Net: Light Convolutional Autoencoder for Image Dehazing
This addresses the need for real-time image pre-processing in low-spec systems, though it appears incremental as it focuses on efficiency rather than a new paradigm.
The paper tackles the problem of computationally inefficient image dehazing by proposing a lightweight convolutional autoencoder that achieves comparable image quality to state-of-the-art methods at a much faster rate on standard datasets.
Image dehazing is a crucial image pre-processing task aimed at removing the incoherent noise generated by haze to improve the visual appeal of the image. The existing models use sophisticated networks and custom loss functions which are computationally inefficient and requires heavy hardware to run. Time is of the essence in image pre-processing since real time outputs can be obtained instantly. To overcome these problems, our proposed generic model uses a very light convolutional encoder-decoder network which does not depend on any atmospheric models. The network complexity-image quality trade off is handled well in this neural network and the performance of this network is not limited by low-spec systems. This network achieves optimum dehazing performance at a much faster rate, on several standard datasets, comparable to the state-of-the-art methods in terms of image quality.