VADER: Video Alignment Differencing and Retrieval
This addresses the spread of misinformation via manipulated videos, providing a tool for verifying video authenticity, though it appears incremental as it builds on existing matching and alignment techniques.
The authors tackled the problem of detecting manipulated videos to combat misinformation by developing VADER, a method for matching, aligning, and summarizing changes in video fragments, which enables robust video provenance analysis.
We propose VADER, a spatio-temporal matching, alignment, and change summarization method to help fight misinformation spread via manipulated videos. VADER matches and coarsely aligns partial video fragments to candidate videos using a robust visual descriptor and scalable search over adaptively chunked video content. A transformer-based alignment module then refines the temporal localization of the query fragment within the matched video. A space-time comparator module identifies regions of manipulation between aligned content, invariant to any changes due to any residual temporal misalignments or artifacts arising from non-editorial changes of the content. Robustly matching video to a trusted source enables conclusions to be drawn on video provenance, enabling informed trust decisions on content encountered.