ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation
نویسندگان
چکیده
We introduce a fast and efficient convolutional neural network, ESPNet, for semantic segmentation of high resolution images under resource constraints. ESPNet is based on a new convolutional module, efficient spatial pyramid (ESP), which is efficient in terms of computation, memory, and power. ESPNet is 22 times faster (on a standard GPU) and 180 times smaller than the stateof-the-art semantic segmentation network PSPNet [1], while its category-wise accuracy is only 8% less. We evaluated EPSNet on a variety of semantic segmentation datasets including Cityscapes, PASCAL VOC, and a breast biopsy whole slide image dataset. Under the same constraints on memory and computation, ESPNet outperforms all the current efficient CNN networks such as MobileNet, ShuffleNet, and ENet on both standard metrics and our newly introduced performance metrics that measure efficiency on edge devices. Our network can process high resolution images at a rate of 112 and 9 frames per second on a standard GPU and edge device, respectively.
منابع مشابه
Multi-Scale Context Aggregation by Dilated Convolutions
State-of-the-art models for semantic segmentation are based on adaptations of convolutional networks that had originally been designed for image classification. However, dense prediction and image classification are structurally different. In this work, we develop a new convolutional network module that is specifically designed for dense prediction. The presented module uses dilated convolution...
متن کاملMultiple Skip Connections and Dilated Convolutions for Semantic Segmentation
We propose a scale-aware semantic segmentation method specifically for small objects. The contributions of this method are 1) to feed the features of a small region by using multiple skip connections, and 2) to extract context from multiple receptive fields by using multiple dilated convolution blocks. The proposed method has achieved high accuracy in the Cityscapes dataset. In comparison with ...
متن کاملLearning Dilation Factors for Semantic Segmentation of Street Scenes
Contextual information is crucial for semantic segmentation. However, finding the optimal trade-off between keeping desired fine details and at the same time providing sufficiently large receptive fields is non trivial. This is even more so, when objects or classes present in an image significantly vary in size. Dilated convolutions have proven valuable for semantic segmentation, because they a...
متن کاملLaplacian Pyramid Reconstruction and Refinement for Semantic Segmentation
CNN architectures have terrific recognition performance but rely on spatial pooling which makes it difficult to adapt them to tasks that require dense, pixel-accurate labeling. This paper makes two contributions: (1) We demonstrate that while the apparent spatial resolution of convolutional feature maps is low, the high-dimensional feature representation contains significant sub-pixel localizat...
متن کاملLaplacian Reconstruction and Refinement for Semantic Segmentation
CNN architectures have terrific recognition performance but rely on spatial pooling which makes it difficult to adapt them to tasks that require dense, pixel-accurate labeling. This paper makes two contributions: (1) We demonstrate that while the apparent spatial resolution of convolutional feature maps is low, the high-dimensional feature representation contains significant sub-pixel localizat...
متن کامل