In this paper, we propose a variety of Long Short-Term Memory (LSTM) based models for sequence tagging. These models include LSTM networks, bidirectional LSTM (BI-LSTM) networks, LSTM with a Conditional Random Field (CRF) layer (LSTM-CRF) and bidirec

ƒ Large format 5 MP CMOS imager (global shutter) ƒ Up to 62 fps at full resolution ƒ 5.0 µm square pixels in a 5:4 aspect ratio ƒ Small size (29 x 29 x 41.5 mm, excluding lens mount) ƒ 8/10/12-bit output in choice of monochrome or raw Bayer color