Abstract | ||
---|---|---|
Given audio of President Barack Obama, we synthesize a high quality video of him speaking with accurate lip sync, composited into a target video clip. Trained on many hours of his weekly address footage, a recurrent neural network learns the mapping from raw audio features to mouth shapes. Given the mouth shape at each time instant, we synthesize high quality mouth texture, and composite it with proper 3D pose matching to change what he appears to be saying in a target video to match the input audio track. Our approach produces photorealistic results. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1145/3072959.3073640 | ACM Trans. Graph. |
Keywords | Field | DocType |
Audio,Face Synthesis,LSTM,RNN,Pig data. Videos,Audiovisual Speech,Uncanny Valley,Lip Sync | Computer vision,Face synthesis,Recurrent neural network,Speech recognition,Raw audio format,Obama,Artificial intelligence,Lip sync,Mathematics,Mouth shape | Journal |
Volume | Issue | ISSN |
36 | 4 | 0730-0301 |
Citations | PageRank | References |
88 | 2.80 | 37 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Supasorn Suwajanakorn | 1 | 266 | 11.20 |
Steven M. Seitz | 2 | 8729 | 495.13 |
Ira Kemelmacher-Shlizerman | 3 | 710 | 28.03 |