Benjamin Kiefer, Lojze Žust, Matej Kristan, Janez Perš, Matija Teršek, Arnold Wiliem, Martin Messmer, Cheng-Yen Yang, Hsiang-Wei Huang, Zhongyu Jiang, Heng-Cheng Kuo, Jie Mei, Jenq-Neng Hwang, Daniel Stadler, Lars Sommer, Kaer Huang, Aiguo Zheng, Weitu Chong, Kanokphan Lertniphonphan, Jun Xie, Feng Chen, Jian Li, Zhepeng Wang, Luca Zedda, Andrea Loddo, Cecilia Di Ruberto, Tuan-Anh Vu, Hai Nguyen-Truong, Tan-Sang Ha, Quan-Dung Pham, Sai-Kit Yeung, Yuan Feng, Nguyen Thanh Thien, Lixin Tian, Sheng-Yao Kuan, Yuan-Hao Ho, Angel Bueno Rodriguez, Borja Carrillo-Perez, Alexander Klein, Antje Alex, Yannik Steiniger, Felix Sattler, Edgardo Solano-Carrillo, Matej Fabijanić, Magdalena Šumunec, Nadir Kapetanović, Andreas Michel, Wolfgang Gross, Martin Weinmann
The 2nd Workshop on Maritime Computer Vision (MaCVi) 2024 addresses maritime computer vision for Unmanned Aerial Vehicles (UAV) and Unmanned Surface Vehicles (USV). Three challenges categories are considered: (i) UAV-based Maritime Object Tracking with Re-identification, (ii) USV-based Maritime Obstacle Segmentation and Detection, (iii) USV-based Maritime Boat Tracking. The USV-based Maritime Obstacle Segmentation and Detection features three sub-challenges, including a new embedded challenge addressing efficicent inference on real-world embedded devices. This report offers a comprehensive overview of the findings from the challenges. We provide both statistical and qualitative analyses, evaluating trends from over 195 submissions. All datasets, evaluation code, and the leaderboard are available to the public at https://macvi.org/workshop/macvi24.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Semantic Segmentation | LaRS | F1 | 79.9 | SWIM^2 (Mask2Former) |
| Semantic Segmentation | LaRS | Q | 78.1 | SWIM^2 (Mask2Former) |
| Semantic Segmentation | LaRS | mIoU | 97.8 | SWIM^2 (Mask2Former) |
| Semantic Segmentation | LaRS | μ | 79.7 | SWIM^2 (Mask2Former) |
| Semantic Segmentation | LaRS | F1 | 80.2 | TransMari (Mask2Former) |
| Semantic Segmentation | LaRS | Q | 77.8 | TransMari (Mask2Former) |
| Semantic Segmentation | LaRS | mIoU | 97.1 | TransMari (Mask2Former) |
| Semantic Segmentation | LaRS | μ | 79.6 | TransMari (Mask2Former) |
| Semantic Segmentation | LaRS | F1 | 77.3 | Mari-Mask2Former (Mask2Former) |
| Semantic Segmentation | LaRS | Q | 75.7 | Mari-Mask2Former (Mask2Former) |
| Semantic Segmentation | LaRS | mIoU | 97.8 | Mari-Mask2Former (Mask2Former) |
| Semantic Segmentation | LaRS | μ | 78.4 | Mari-Mask2Former (Mask2Former) |
| Semantic Segmentation | LaRS | F1 | 58.9 | eWaSR |
| Semantic Segmentation | LaRS | Q | 56.5 | eWaSR |
| Semantic Segmentation | LaRS | mIoU | 96 | eWaSR |
| Semantic Segmentation | LaRS | μ | 67.8 | eWaSR |
| 10-shot image generation | LaRS | F1 | 79.9 | SWIM^2 (Mask2Former) |
| 10-shot image generation | LaRS | Q | 78.1 | SWIM^2 (Mask2Former) |
| 10-shot image generation | LaRS | mIoU | 97.8 | SWIM^2 (Mask2Former) |
| 10-shot image generation | LaRS | μ | 79.7 | SWIM^2 (Mask2Former) |
| 10-shot image generation | LaRS | F1 | 80.2 | TransMari (Mask2Former) |
| 10-shot image generation | LaRS | Q | 77.8 | TransMari (Mask2Former) |
| 10-shot image generation | LaRS | mIoU | 97.1 | TransMari (Mask2Former) |
| 10-shot image generation | LaRS | μ | 79.6 | TransMari (Mask2Former) |
| 10-shot image generation | LaRS | F1 | 77.3 | Mari-Mask2Former (Mask2Former) |
| 10-shot image generation | LaRS | Q | 75.7 | Mari-Mask2Former (Mask2Former) |
| 10-shot image generation | LaRS | mIoU | 97.8 | Mari-Mask2Former (Mask2Former) |
| 10-shot image generation | LaRS | μ | 78.4 | Mari-Mask2Former (Mask2Former) |
| 10-shot image generation | LaRS | F1 | 58.9 | eWaSR |
| 10-shot image generation | LaRS | Q | 56.5 | eWaSR |
| 10-shot image generation | LaRS | mIoU | 96 | eWaSR |
| 10-shot image generation | LaRS | μ | 67.8 | eWaSR |