Learning Large-scale Neural Fields via Context Pruned Meta-Learning

We introduce an efficient optimization-based meta-learning technique for large-scale neural field training by realizing significant memory savings through automated online context point selection.