Attention based Memory video portrait matting
arXiv:2203.06890v2h-index: 2
Originality Incremental advance
AI Analysis
This addresses video matting for portrait editing, but appears incremental as it builds on existing attention mechanisms.
The paper tackled video portrait matting without trimaps by using an attention-based method, resulting in a novel approach that avoids computationally expensive modules.
We proposed a novel trimap free video matting method based on the attention mechanism. By the nature of the problem, most existing approaches use either multiple computational expansive modules or complex algorithms to exploit temporal information fully. We designed a temporal aggregation module to compute the temporal coherence between the current frame and its two previous frames.