We introduce O-SegNet- the robust encoder and decoder architecture for objects segmentation from high-resolution aerial imagery data to address this challenge. The proposed O-SegNet architecture ...

Impact Statement: Recognizing human actions in dark-light conditions has many real-time applications like night-smart surveillance systems, elderly people, monitoring in smart homes, military ...

there is no adversarial network, nor any image loss, robustness with regards to JPEG is done by stopping the gradient (see Towards Robust Deep Hiding Under Non-Differentiable Distortions for Practical ...