CondNet: Conditional Classifier for Scene Segmentation

21 Sep 2021  ·  Changqian Yu, Yuanjie Shao, Changxin Gao, Nong Sang ·

The fully convolutional network (FCN) has achieved tremendous success in dense visual recognition tasks, such as scene segmentation. The last layer of FCN is typically a global classifier (1x1 convolution) to recognize each pixel to a semantic label. We empirically show that this global classifier, ignoring the intra-class distinction, may lead to sub-optimal results. In this work, we present a conditional classifier to replace the traditional global classifier, where the kernels of the classifier are generated dynamically conditioned on the input. The main advantages of the new classifier consist of: (i) it attends on the intra-class distinction, leading to stronger dense recognition capability; (ii) the conditional classifier is simple and flexible to be integrated into almost arbitrary FCN architectures to improve the prediction. Extensive experiments demonstrate that the proposed classifier performs favourably against the traditional classifier on the FCN architecture. The framework equipped with the conditional classifier (called CondNet) achieves new state-of-the-art performances on two datasets. The code and models are available at https://git.io/CondNet.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Semantic Segmentation ADE20K CondNet(ResNet-101) Validation mIoU 47.38 # 156
Semantic Segmentation ADE20K CondNet(ResNest-101) Validation mIoU 47.54 # 154
Semantic Segmentation PASCAL Context CondNet(ResNest-101) mIoU 57 # 18
Semantic Segmentation PASCAL Context CondNet(ResNet-101) mIoU 56.0 # 23

Methods