Abstract: Task-oriented video compression aims to eliminate redundancy while preserving task-critical information. However, existing spatial domain methods incur high computational overhead, whereas ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results