Multi-Grained Spatio-temporal Modeling for Lip-reading
About
Lip-reading aims to recognize speech content from videos via visual analysis of speakers' lip movements. This is a challenging task due to the existence of homophemes-words which involve identical or highly similar lip movements, as well as diverse lip appearances and motion patterns among the speakers. To address these challenges, we propose a novel lip-reading model which captures not only the nuance between words but also styles of different speakers, by a multi-grained spatio-temporal modeling of the speaking process. Specifically, we first extract both frame-level fine-grained features and short-term medium-grained features by the visual front-end, which are then combined to obtain discriminative representations for words with similar phonemes. Next, a bidirectional ConvLSTM augmented with temporal attention aggregates spatio-temporal information in the entire input sequence, which is expected to be able to capture the coarse-gained patterns of each word and robust to various conditions in speaker identity, lighting conditions, and so on. By making full use of the information from different levels in a unified framework, the model is not only able to distinguish words with similar pronunciations, but also becomes robust to appearance changes. We evaluate our method on two challenging word-level lip-reading benchmarks and show the effectiveness of the proposed method, which also demonstrate the above claims.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Lip-reading | LRW-1000 (test) | Accuracy36.91 | 50 | |
| Lip-reading Classification | LRW (test) | Accuracy83.34 | 38 | |
| Word Recognition | LRW (test) | Correct Rate83.34 | 13 | |
| Lip-reading | LRW Word-level (test) | Accuracy83.3 | 13 | |
| Lip-reading Classification | LRW-1000 cropped mouth regions (test) | Top-1 Accuracy0.369 | 9 | |
| Lip-reading | LRW (test) | Word Accuracy83.3 | 8 | |
| Word-level lip reading | LRW-1000 | Accuracy0.3691 | 4 |