watermark-removal and watermark-remover
These are competitors offering alternative approaches to watermark removal—one uses deep learning-based inpainting for arbitrary watermark positions while the other uses traditional image processing for fixed-position watermarks.
About watermark-removal
zuruoke/watermark-removal
a machine learning image inpainting task that instinctively removes watermarks from image indistinguishable from the ground truth image
Implements contextual attention and gated convolution architectures from CVPR 2018 and ICCV 2019 papers to reconstruct missing image regions. Built on TensorFlow 1.15.0 with the neuralgym toolkit, supporting multiple watermark types (e.g., iStock) and deployable via Docker for batch processing. Accepts watermarked images and outputs reconstructed versions through a command-line interface with configurable checkpoint directories.
About watermark-remover
whitelok/watermark-remover
Remove watermark automatically(Just can use for fixed position watermark till now). 自动水印消除算法的实现（目前只支持固定水印位置）。
Scores updated daily from GitHub, PyPI, and npm data. How scores work