Robust High-Resolution Video Matting with Temporal Guidance

25 Aug 2021  ·  Shanchuan Lin, Linjie Yang, Imran Saleemi, Soumyadip Sengupta ·

We introduce a robust, real-time, high-resolution human video matting method that achieves new state-of-the-art performance. Our method is much lighter than previous approaches and can process 4K at 76 FPS and HD at 104 FPS on an Nvidia GTX 1080Ti GPU. Unlike most existing methods that perform video matting frame-by-frame as independent images, our method uses a recurrent architecture to exploit temporal information in videos and achieves significant improvements in temporal coherence and matting quality. Furthermore, we propose a novel training strategy that enforces our network on both matting and segmentation objectives. This significantly improves our model's robustness. Our method does not require any auxiliary inputs such as a trimap or a pre-captured background image, so it can be widely applied to existing human matting applications.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Video Matting ImageNet Lin et al. 2021 Alpha - MAD 6.08 # 1
Alpha - MSE 1.47 # 1
Alpha - Grad 0.88 # 1
Alpha - Conn 0.41 # 1
Alpha - dtSSD 1.36 # 1
Video Matting ImageNet MODNet Alpha - MAD 9.41 # 2
Alpha - MSE 4.30 # 2
Alpha - Grad 1.89 # 2
Alpha - Conn 0.81 # 2
Alpha - dtSSD 2.23 # 2


No methods listed for this paper. Add relevant methods here