What is the central idea behind deep-learning-based super-resolution techniques (such as DLSS) used in real-time rendering?
Correct! Well done.
Incorrect.
The correct answer is C) A trained neural network upscales and reconstructs a sharper, higher-resolution frame from a lower-resolution rendered input — often combined with temporal data — cutting the rendering workload while preserving perceived quality
Correct Answer
A trained neural network upscales and reconstructs a sharper, higher-resolution frame from a lower-resolution rendered input — often combined with temporal data — cutting the rendering workload while preserving perceived quality
Neural super-resolution renders internally at a lower (cheaper) resolution and uses a trained model — frequently leveraging temporal information — to reconstruct a sharper, higher-resolution output, yielding large performance gains with minimal perceptible quality loss.