WaSR—A Water Segmentation and Refinement Maritime Obstacle Detection Network

Bovcon, Borja; Kristan, Matej

doi:10.1109/tcyb.2021.3085856

Cited by 51 publications

(55 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…shore, piers, floating fences) in a unified way by posing the problem as anomaly segmentation. Recently, several specialized networks for the marine domain have been proposed for this task [5], [7], [9]. These methods address reflections and increase detection robustness in multiple ways, including regularization techniques [7], specialized loss functions [5] and obstacle-oriented training regimes [8].…”

Section: Related Workmentioning

confidence: 99%

“…We propose using the temporal context to improve the prediction accuracy. Our network (Figure 2), denoted as WaSR-T, is based on the state-ofthe-art single-frame network for maritime obstacle detection WaSR [5]. We design WaSR-T to encode the discriminative temporal information about local appearance changes of a region over T preceding context frames M ∈ R T ×3×H×W .…”

Section: Temporal Context For Obstacle Detectionmentioning

confidence: 99%

“…However, recent works [3], [4] have shown that methods developed for autonomous cars do not translate well to USVs due to the specifics of the maritime domain. As a result, several approaches that exploit the domain specifics for improved detection accuracy have been recently proposed [5], [6], [7], [8]. Since everything but water can be an obstacle, classical detectors for individual obstacle classes cannot address all obstacle types.…”

Section: Introductionmentioning

confidence: 99%

“…Since everything but water can be an obstacle, classical detectors for individual obstacle classes cannot address all obstacle types. State-ofthe-art methods [5] instead casts maritime obstacle detection as an anomaly segmentation problem by segmenting the image into the water, sky and obstacle categories.…”

Section: Introductionmentioning

confidence: 99%

“…Results on the recent maritime benchmark MODS [4] show that, compared to the single-frame WaSR [5], the proposed WaSR-T reduces the number of false positive detections by 30% with a low computational overhead and sets a new state-of-the-art in maritime obstacle detection. In summary, our main contributions are: (i) WaSR-T, a temporal extension of WaSR [5] that leverages the temporal context for increased robustness and (ii) MaSTr1478, an extension of the existing single-frame training dataset [3] with challenging reflection scenes that facilitates the training of temporal maritime segmentation networks. The new dataset and the WaSR-T source code will be publicly released to facilitate further research of temporal features in maritime obstacle detection.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Temporal Context for Robust Maritime Obstacle Detection

Žust¹,

Kristan²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Robust maritime obstacle detection is essential for fully autonomous unmanned surface vehicles (USVs). The currently widely adopted segmentation-based obstacle detection methods are prone to misclassification of object reflections and sun glitter as obstacles, producing many false positive detections, effectively rendering the methods impractical for USV navigation. However, water-turbulence-induced temporal appearance changes on object reflections are very distinctive from the appearance dynamics of true objects. We harness this property to design WaSR-T, a novel maritime obstacle detection network, that extracts the temporal context from a sequence of recent frames to reduce ambiguity. By learning the local temporal characteristics of object reflection on the water surface, WaSR-T substantially improves obstacle detection accuracy in the presence of reflections and glitter. Compared with existing single-frame methods, WaSR-T reduces the number of false positive detections by 41% overall and by over 53% within the danger zone of the boat, while preserving a high recall, and achieving new state-of-the-art performance on the challenging MODS maritime obstacle detection benchmark.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Temporal Context For Obstacle Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Temporal Context for Robust Maritime Obstacle Detection

Žust¹,

Kristan²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A method for detecting navigable areas in narrow rivers under complex reflection conditions

Zhang,

Hu,

et al. 2024

Engineering Reports

View full text Add to dashboard Cite

The perception of unmanned surface vehicles is significantly influenced by the detection of navigable areas in narrow rivers. Conventional semantic segmentation networks are unable to resolve the numerous interferences on the water's surface, including highlights and inverted images. To solve this problem, a river surface image reflection removal generative adversarial network (RRGAN) is proposed to eliminate the interference of harsh water surface environment. The proposed RRGAN only uses a single generator to reduce the number of parameters. By adding AdaLIN layers in the generator to enhance the ability to generate low‐reflection images, the AdaLIN encoder (AdaLINE) is proposed to automatically generate normalized affine parameters. In addition, a cycle semantic consistency loss function with a single generator is proposed to ensure that the water region of the generated images remains unchanged. Finally, a two‐stage method for detecting navigable areas is proposed. In the first stage, the RRGAN is used to remove the interference on the water surface environment. In the second stage, the semantic segmentation network is used to segment the water body from the denoised image to determine the navigable areas on the water surface. The experimental results demonstrate that, in the complex and varied narrow river environment, the suggested RRGAN method can significantly reduce the reflection interference of the water surface and improve the accuracy of the water segmentation after the reflection is removed.

show abstract