Efficient Yet Deep Convolutional Neural Networks for Semantic Segmentation

Semantic Segmentation using deep convolutional neural network pose more complex challenge for any GPU intensive task. As it has to compute million of parameters, it results to huge memory consumption... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Semantic Segmentation NYU Depth v2 Dilated FCN-2s RGB Mean IoU 32.3% # 10
Scene Segmentation NYU Depth v2 Dilated FCN-2s RGB Mean IoU 32.3% # 1
Semantic Segmentation PASCAL Context Dilated-FCN2s mIoU 42.6 # 37
Semantic Segmentation PASCAL VOC 2012 test Dilated FCN-2s VGG19 Mean IoU 69% # 43

Methods used in the Paper


METHOD TYPE
Dilated Convolution
Convolutions
Convolution
Convolutions