1^st Workshop on Maritime Computer Vision (MaCVi) 2023: Challenge Results

Abstract

The 1 st Workshop on Maritime Computer Vision (MaCVi) 2023 focused on maritime computer vision for Unmanned Aerial Vehicles (UAV) and Unmanned Surface Vehicle (USV), and organized several subchallenges in this domain: (i) UAV-based Maritime Object Detection, (ii) UAV-based Maritime Object Tracking, (iii) USV-based Maritime Obstacle Segmentation and (iv) USV-based Maritime Obstacle Detection. The subchallenges were based on the SeaDronesSee and MODS benchmarks. This report summarizes the main findings of the individual subchallenges and introduces a new benchmark, called SeaDronesSee Object Detection v2, which extends the previous benchmark by including more classes and footage. We provide statistical and qualitative analyses, and assess trends in the best-performing methodologies of over 130 submissions. The methods are summarized in the appendix. The datasets, evaluation code and the leaderboard are publicly available (https:// seadronessee.cs.uni-tuebingen.de/macvi).

Diminishing Domain Bias by Leveraging Domain Labels in Object Detection on UAVs

“…We made sure to have a great -90 • ). Note that we focus on the detection of people in both benchmarks only, since this is the hardest class to detect [14]. For that, we fuse the classes swimmers, swimmers with life jacket and life jacket into a single people class in SeaDronesSee-V (SeaDronesSee-MOT without instance IDs).…”

“…To provide evidence for the initial claim about the high recall before NMS, we train a Faster R-CNN ResNet-18 and a YOLOv7 (configs from [14]) on SeaDronesSee v2 [14]. For evaluation, we remove the NMS stage and obtain recall values of 97.4 % and 98.8%, resp.…”

On Automatic Person-in-Water Detection for Marine Search and Rescue Operations

“…SeaDroneSee dataset [36] includes 54,000 frames with humans in open water captured with drones from various altitudes and viewing angles ranging from 5 to 260 meters and 0 to 90 degrees while providing the respective meta information for altitude, viewing angle and other metadata. A Maritime Computer Vision Challenge [37] included tasks for UAV-based maritime object detection and tracking. An approach for obtaining fast region-of-interest proposals in a video stream on an embedded GPU for maritime human detection tasks was proposed in [38].…”

POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments

“…The selected dataset for experimentation was the SeaDronesSee Object Detection v2 [ 8 ], which comprises 8931 images and 57,760 instances in the training set. Notably, this dataset is highly imbalanced, as previously observed.…”