# Training Constrained Deconvolutional Networks for Road Scene Semantic Segmentation

arXiv: https://arxiv.org/abs/1604.01545 · HTML: https://mlsanity.com/paper/1604.01545/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Apr 2016.

## Built on by (most cited first, 13 of 13)

- [MSeg: A Composite Dataset for Multi-domain Semantic Segmentation](https://mlsanity.com/paper/2112.13762/) (2021)
- [Synthetic Datasets for Autonomous Driving: A Survey](https://mlsanity.com/paper/2304.12205/) (2023)
- [COLA: COarse-LAbel multi-source LiDAR semantic segmentation for autonomous driving](https://mlsanity.com/paper/2311.03017/) (2023)
- [A Systematic Review of Generalization Research in Medical Image Classification](https://mlsanity.com/paper/2403.12167/) (2024)
- [Scaling up Multi-domain Semantic Segmentation with Sentence Embeddings](https://mlsanity.com/paper/2202.02002/) (2022)
- [COLA: COarse LAbel pre-training for 3D semantic segmentation of sparse LiDAR datasets](https://mlsanity.com/paper/2202.06884/) (2022)
- [Seamless Iterative Semi-Supervised Correction of Imperfect Labels in Microscopy Images](https://mlsanity.com/paper/2208.03327/) (2022)
- [An Empirical Study on Multi-Domain Robust Semantic Segmentation](https://mlsanity.com/paper/2212.04221/) (2022)
- [Training Semantic Segmentation on Heterogeneous Datasets](https://mlsanity.com/paper/2301.07634/) (2023)
- [MINERVAS: Massive INterior EnviRonments VirtuAl Synthesis](https://mlsanity.com/paper/2107.06149/) (2021)
- [Deep Learning Techniques for Visual Counting](https://mlsanity.com/paper/2206.03033/) (2022)
- [Enhancing Object Detection Accuracy in Autonomous Vehicles Using Synthetic Data](https://mlsanity.com/paper/2411.15602/) (2024)
- [AMA-SAM: Adversarial Multi-Domain Alignment of Segment Anything Model for High-Fidelity Histology Nuclei Segmentation](https://mlsanity.com/paper/2503.21695/) (2025)

## Abstract (arXiv)

In this work we investigate the problem of road scene semantic segmentation using Deconvolutional Networks (DNs). Several constraints limit the practical performance of DNs in this context: firstly, the paucity of existing pixel-wise labelled training data, and secondly, the memory constraints of embedded hardware, which rule out the practical use of state-of-the-art DN architectures such as fully convolutional networks (FCN). To address the first constraint, we introduce a Multi-Domain Road Scene Semantic Segmentation (MDRS3) dataset, aggregating data from six existing densely and sparsely labelled datasets for training our models, and two existing, separate datasets for testing their generalisation performance. We show that, while MDRS3 offers a greater volume and variety of data, end-to-end training of a memory efficient DN does not yield satisfactory performance. We propose a new training strategy to overcome this, based on (i) the creation of a best-possible source network (S-Net) from the aggregated data, ignoring time and memory constraints; and (ii) the transfer of knowledge from S-Net to the memory-efficient target network (T-Net). We evaluate different techniques for S-Net creation and T-Net transferral, and demonstrate that training a constrained deconvolutional network in this manner can unlock better performance than existing training approaches. Specifically, we show that a target network can be trained to achieve improved accuracy versus an FCN despite using less than 1% of the memory. We believe that our approach can be useful beyond automotive scenarios where labelled data is similarly scarce or fragmented and where practical constraints exist on the desired model size. We make available our network models and aggregated multi-domain dataset for reproducibility.