DiCENet: Dimension-wise Convolutions for Efficient Networks

8 Jun 2019Sachin MehtaHannaneh HajishirziMohammad Rastegari

In this paper, we propose a new CNN model DiCENet, that is built using: (1) dimension-wise convolutions and (2) efficient channel fusion. The introduced blocks maximize the use of information in the input tensor by learning representations across all dimensions while simultaneously reducing the complexity of the network and achieving high accuracy... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Semantic Segmentation Cityscapes DiCENet Mean IoU (class) 63.4% # 25
Image Classification ImageNet DiCENet Top 1 Accuracy 75.1% # 72
Object Detection PASCAL VOC 2007 DiCENet MAP 65.2% # 20