LRW-1000 is a naturally-distributed large-scale benchmark for word-level lipreading in the wild, including 1000 classes with about 718,018 video samples from more than 2000 individual speakers. There are more than 1,000,000 Chinese character instances in total. Each class corresponds to the syllables of a Mandarin word which is composed by one or several Chinese characters. This dataset aims to cover a natural variability over different speech modes and imaging conditions to incorporate challenges encountered in practical applications. It shows a large variation over several aspects, including the number of samples in each class, resolution of videos, lighting conditions, and speakers' attributes such as pose, age, gender, and make-up and so on, as shown in Fig. 1 and Fig. 2.
* Note that LRW-1000 has been renamed as CAS-VSR-W1k. You may refer to it as “CAS-VSR-W1k (The original LRW-1000)”.
Fig.1 The diversity of the speakers' appearance in CAS-VSR-W1k (the original LRW-1000)
Fig.2 Lip Samples in CAS-VSR-W1k (the original LRW-1000)
718,018 samples with an average of 718 samples for each class
1000 classes, with each class corresponds to the syllables of a Mandarin word
~2000 different speakers with a large coverage over speech modes, including speech rate, viewpoint, age, gender and make-up and so on
3. Evaluation Protocols
We provide two evaluation metrics for experiments. A). The recognition accuracy over all 1000 classes is naturally considered as the base metric, since this is a classification task. B). Motivated by the large diversity of the data shown in many aspects, such as the number of samples in each class, we also provide the Kappa Coefficient as a second evaluation metric.
The database is public to universities and research institutes for research purpose only. To request a copy of the database, please do as follows:
5. Contact Info