Fast Single Image Reflection Suppression via Convex Optimization
This addresses the need for fast and effective reflection suppression in computer vision applications, such as image enhancement and preprocessing for machine learning, representing an incremental improvement in efficiency.
The paper tackles the problem of removing undesired reflections from images taken through glass by proposing a convex optimization model, which achieves desirable suppression results and dramatically reduces execution time compared to previous methods.
Removing undesired reflections from images taken through the glass is of great importance in computer vision. It serves as a means to enhance the image quality for aesthetic purposes as well as to preprocess images in machine learning and pattern recognition applications. We propose a convex model to suppress the reflection from a single input image. Our model implies a partial differential equation with gradient thresholding, which is solved efficiently using Discrete Cosine Transform. Extensive experiments on synthetic and real-world images demonstrate that our approach achieves desirable reflection suppression results and dramatically reduces the execution time.