The DEVIL is in the Details: A Diagnostic Evaluation Benchmark for Video Inpainting

TL;DR

The paper introduces DEVIL, a benchmark with a new dataset and evaluation scheme for video inpainting, focusing on failure modes and content attributes to better diagnose method performance.

Contribution

It presents a novel dataset with labeled failure modes and an evaluation scheme that isolates content attributes, enabling detailed analysis of video inpainting methods.

Findings

Reveals how specific content attributes affect inpainting performance.

Provides a diagnostic tool for systematic analysis of failure modes.

Enables targeted improvements in video inpainting techniques.

Abstract

Quantitative evaluation has increased dramatically among recent video inpainting work, but the video and mask content used to gauge performance has received relatively little attention. Although attributes such as camera and background scene motion inherently change the difficulty of the task and affect methods differently, existing evaluation schemes fail to control for them, thereby providing minimal insight into inpainting failure modes. To address this gap, we propose the Diagnostic Evaluation of Video Inpainting on Landscapes (DEVIL) benchmark, which consists of two contributions: (i) a novel dataset of videos and masks labeled according to several key inpainting failure modes, and (ii) an evaluation scheme that samples slices of the dataset characterized by a fixed content attribute, and scores performance on each slice according to reconstruction, realism, and temporal consistency quality. By revealing systematic changes in performance induced by particular characteristics of the input content, our challenging benchmark enables more insightful analysis into video inpainting methods and serves as an invaluable diagnostic tool for the field. Our code and data are available at https://github.com/MichiganCOG/devil .