# Dilated Continuous Random Field for Semantic Segmentation

arXiv: https://arxiv.org/abs/2202.00162 · HTML: https://mlsanity.com/paper/2202.00162/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 23 references that are in the map.
- First posted on arXiv 1 Feb 2022.

## Builds on (most cited first, 11 of 11)

- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Conditional Random Fields as Recurrent Neural Networks](https://mlsanity.com/paper/1502.03240/) (2015)
- [Efficient piecewise training of deep structured models for semantic segmentation](https://mlsanity.com/paper/1504.01013/) (2015)
- [Self-Supervised Difference Detection for Weakly-Supervised Semantic Segmentation](https://mlsanity.com/paper/1911.01370/) (2019)
- [Learning Discriminative Feature with CRF for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2008.01270/) (2020)
- [Fractional Max-Pooling](https://mlsanity.com/paper/1412.6071/) (2014)
- [Convolutional CRFs for Semantic Segmentation](https://mlsanity.com/paper/1805.04777/) (2018)
- [A Multi-scale CNN-CRF Framework for Environmental Microorganism Image Segmentation](https://mlsanity.com/paper/2003.03744/) (2020)
- [Superpixel-enhanced Pairwise Conditional Random Field for Semantic Segmentation](https://mlsanity.com/paper/1805.11737/) (2018)

## Abstract (arXiv)

Mean field approximation methodology has laid the foundation of modern Continuous Random Field (CRF) based solutions for the refinement of semantic segmentation. In this paper, we propose to relax the hard constraint of mean field approximation - minimizing the energy term of each node from probabilistic graphical model, by a global optimization with the proposed dilated sparse convolution module (DSConv). In addition, adaptive global average-pooling and adaptive global max-pooling are implemented as replacements of fully connected layers. In order to integrate DSConv, we design an end-to-end, time-efficient DilatedCRF pipeline. The unary energy term is derived either from pre-softmax and post-softmax features, or the predicted affordance map using a conventional classifier, making it easier to implement DilatedCRF for varieties of classifiers. We also present superior experimental results of proposed approach on the suction dataset comparing to other CRF-based approaches.