Timezone: »

Two-Stream Network for Sign Language Recognition and Translation
Yutong Chen · Ronglai Zuo · Fangyun Wei · Yu Wu · Shujie LIU · Brian Mak

Thu Dec 01 02:00 PM -- 04:00 PM (PST) @ Hall J #136

Sign languages are visual languages using manual articulations and non-manual elements to convey information. For sign language recognition and translation, the majority of existing approaches directly encode RGB videos into hidden representations. RGB videos, however, are raw signals with substantial visual redundancy, leading the encoder to overlook the key information for sign language understanding. To mitigate this problem and better incorporate domain knowledge, such as handshape and body movement, we introduce a dual visual encoder containing two separate streams to model both the raw videos and the keypoint sequences generated by an off-the-shelf keypoint estimator. To make the two streams interact with each other, we explore a variety of techniques, including bidirectional lateral connection, sign pyramid network with auxiliary supervision, and frame-level self-distillation. The resulting model is called TwoStream-SLR, which is competent for sign language recognition (SLR). TwoStream-SLR is extended to a sign language translation (SLT) model, TwoStream-SLT, by simply attaching an extra translation network. Experimentally, our TwoStream-SLR and TwoStream-SLT achieve state-of-the-art performance on SLR and SLT tasks across a series of datasets including Phoenix-2014, Phoenix-2014T, and CSL-Daily.

Author Information

Yutong Chen (Tsinghua University, Tsinghua University)
Ronglai Zuo (The Hong Kong University of Science and Technology)
Fangyun Wei (Microsoft Research Asia)
Yu Wu (Microsoft)
Shujie LIU (Microsoft)
Brian Mak (The Hong Kong University of Science and Technology)

More from the Same Authors