# Learning Maritime Obstacle Detection from Weak Annotations by Scaffolding

arXiv: https://arxiv.org/abs/2108.00564 · HTML: https://mlsanity.com/paper/2108.00564/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 46 references that are in the map.
- First posted on arXiv 1 Aug 2021.

## Builds on (most cited first, 30 of 30)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Pyramid Scene Parsing Network](https://mlsanity.com/paper/1612.01105/) (2016)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation](https://mlsanity.com/paper/1808.00897/) (2018)
- [RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation](https://mlsanity.com/paper/1611.06612/) (2016)
- [ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1604.05144/) (2016)
- [Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1803.10464/) (2018)
- [Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2004.04581/) (2020)
- [What's the Point: Semantic Segmentation with Point Supervision](https://mlsanity.com/paper/1506.02106/) (2015)
- [BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1503.01640/) (2015)
- [Simple Does It: Weakly Supervised Instance and Semantic Segmentation](https://mlsanity.com/paper/1603.07485/) (2016)
- [Adversarial Learning for Semi-Supervised Semantic Segmentation](https://mlsanity.com/paper/1802.07934/) (2018)
- [BoxInst: High-Performance Instance Segmentation with Box Annotations](https://mlsanity.com/paper/2012.02310/) (2020)
- [Semi-Supervised Semantic Segmentation with High- and Low-level Consistency](https://mlsanity.com/paper/1908.05724/) (2019)
- [Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation](https://mlsanity.com/paper/1805.04574/) (2018)
- [Deep Extreme Cut: From Extreme Points to Object Segmentation](https://mlsanity.com/paper/1711.09081/) (2017)
- [Learning random-walk label propagation for weakly-supervised semantic segmentation](https://mlsanity.com/paper/1802.00470/) (2018)
- [Weakly-Supervised Salient Object Detection via Scribble Annotations](https://mlsanity.com/paper/2003.07685/) (2020)
- [Learning What to Learn for Video Object Segmentation](https://mlsanity.com/paper/2003.11540/) (2020)
- [Universal Semi-Supervised Semantic Segmentation](https://mlsanity.com/paper/1811.10323/) (2018)
- [Video Processing from Electro-optical Sensors for Object Detection and Tracking in Maritime Environment: A Survey](https://mlsanity.com/paper/1611.05842/) (2016)
- [MODS -- A USV-oriented object detection and obstacle segmentation benchmark](https://mlsanity.com/paper/2105.02359/) (2021)
- [Fast image-based obstacle detection from unmanned surface vehicles](https://mlsanity.com/paper/1503.01918/) (2015)
- [Stereo obstacle detection for unmanned surface vehicles by IMU-assisted semantic segmentation](https://mlsanity.com/paper/1802.07956/) (2018)
- [Finding Berries: Segmentation and Counting of Cranberries using Point Supervision and Shape Priors](https://mlsanity.com/paper/2004.08501/) (2020)
- [Generating Masks from Boxes by Mining Spatio-Temporal Consistencies in Videos](https://mlsanity.com/paper/2101.02196/) (2021)
- [A water-obstacle separation and refinement network for unmanned surface vehicles](https://mlsanity.com/paper/2001.01921/) (2020)

## Built on by (most cited first, 8 of 8)

- [A Guide to Image and Video based Small Object Detection using Deep Learning : Case Study of Maritime Surveillance](https://mlsanity.com/paper/2207.12926/) (2022)
- [Temporal Context for Robust Maritime Obstacle Detection](https://mlsanity.com/paper/2203.05352/) (2022)
- [3rd Workshop on Maritime Computer Vision (MaCVi) 2025: Challenge Results](https://mlsanity.com/paper/2501.10343/) (2025)
- [Learning with Weak Annotations for Robust Maritime Obstacle Detection](https://mlsanity.com/paper/2206.13263/) (2022)
- [Improving the Detection of Small Oriented Objects in Aerial Images](https://mlsanity.com/paper/2401.06503/) (2024)
- [Quality-Driven and Diversity-Aware Sample Expansion for Robust Marine Obstacle Segmentation](https://mlsanity.com/paper/2512.13970/) (2025)
- [4th Workshop on Maritime Computer Vision (MaCVi): Challenge Overview](https://mlsanity.com/paper/2604.13244/) (2026)
- [PAANI : On Device Visual Evidence Fusion and Explainable Guidance for River Robot Simulation](https://mlsanity.com/paper/2609.22353/) (2026)

## Abstract (arXiv)

Coastal water autonomous boats rely on robust perception methods for obstacle detection and timely collision avoidance. The current state-of-the-art is based on deep segmentation networks trained on large datasets. Per-pixel ground truth labeling of such datasets, however, is labor-intensive and expensive. We observe that far less information is required for practical obstacle avoidance - the location of water edge on static obstacles like shore and approximate location and bounds of dynamic obstacles in the water is sufficient to plan a reaction. We propose a new scaffolding learning regime (SLR) that allows training obstacle detection segmentation networks only from such weak annotations, thus significantly reducing the cost of ground-truth labeling. Experiments show that maritime obstacle segmentation networks trained using SLR substantially outperform the same networks trained with dense ground truth labels. Thus accuracy is not sacrificed for labelling simplicity but is in fact improved, which is a remarkable result.