In this paper, we design a convolutional neural network based on the ideas of depthwise separable convolution and inverted residual module. The scaling factor of BN layer is used as a measure for channel pruning of the network model to compress it. By analyzing the layer-by-layer pruning process of conventional convolution, the layer-by-layer pruning method with depthwise separable convolution and inverted residual structure is proposed to prune the channels of the network model, and finally, the channel pruning strategy of classification simplification network is developed. Tests on the selected dataset showed that the classification accuracy of the pruned and fine-tuned network model is 97.7% when the pruning rate is 0.7.
Monocular vehicle perception has been a vital problem in autonomous driving. Anchor-based detectors are always used to solve the problem but the used bounding box’s center to calibrate is actually not corresponded with vehicle’s 3D center. Therefore, we designed a perception network by point detector U-Net and propose a calibration method by using the 3D2D reprojection constraints and a prior plane in the imaging system. We evaluate our method on the ApolloCar3D dataset, and the experimental results prove that the proposed network is feasible and the calibration is effective. On the ApolloCar3D testing dataset, the mean average precision of the proposed network is 0.022, and can be well improved to 0.040 with our calibration method, which is competitive compared to other algorithms.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.