As an important part of autonomous driving intelligence perception, pedestrian detection has high requirements for parameter size, real-time, and model performance. Firstly, a novel multiplexed connection residual block is proposed to construct the lightweight network improving ability extract features. Secondly, scalable attention module investigated expand local perceptual field based on dila...

Conventional zero-shot learning aims to train a classifier on training set (seen classes) recognize instances of novel classes (unseen by class-level semantic attributes. In generalized (GZSL), the needs both seen and unseen classes, which is problem extreme data imbalance. To solve this problem, feature generative methods have been proposed make up for lack classes. Current use class attribute...

In this paper, we propose a deep residual dense network (DRDN) for single image super- resolution. Based on human perceptual characteristics, the in block strategy (RRDB) is exploited to implement various depths architectures. The proposed model exhibits simple sequential structure comprising and blocks with skip connections. It improves stability computational complexity of network, as well qu...

The atmosphere in low latitude regions is of particular interest to GPS researchers because the propagation of GPS signals becomes significantly delayed compared with other regions of the world. Hence this limits GPS positioning accuracy in equatorial regions. Although the atmospheric delay can be modelled, a residual component will still remain. Reducing, or mitigating the effect of residual a...