This special issue comprises four papers that contribute to computer vision from various aspects, including automated understanding of funny moments in videos, measuring the similarity of 3D skeleton action sequences, cross-architecture knowledge distillation, and selective classification with out-of-distribution data.

As program chairs of ACCV 2022 and guest editors of this special issue, we invited authors of the award-winning papers to submit extended manuscripts. These manuscripts underwent thorough peer review according to journal standards before being accepted for inclusion in this special issue.

In our capacity as guest editors, we express our gratitude to all authors for their significant contributions to this special issue. We also extend our appreciation to the reviewers whose insightful feedback further enhanced the quality of the papers, as well as to the editorial team at IJCV for their invaluable support. It is our hope that the papers featured in this special issue will serve as a valuable resource for both researchers and practitioners in the field of computer vision.