Multi-Task Learning (MTL) aims to enhance the model generalization by sharing representations between related tasks for better performance. Typical MTL methods are jointly trained with the complete multitude of ground-truths for all tasks simultaneously. However, one single dataset may not contain the annotations for each task of interest. To address this issue, we propose the Semi-supervised Multi-Task Learning (SemiMTL) method to leverage the available supervisory signals from different datasets, particularly for semantic segmentation and depth estimation tasks. To this end, we design an adversarial learning scheme in our semi-supervised training by leveraging unlabeled data to optimize all the task branches simultaneously and accomplish all tasks across datasets with partial annotations. We further present a domain-aware discriminator structure with various alignment formulations to mitigate the domain discrepancy issue among datasets. Finally, we demonstrate the effectiveness of the proposed method to learn across different datasets on challenging street view and remote sensing benchmarks.
|Title of host publication||Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Number of pages||10|
|Publication status||Published - 2022|
|Event||22nd IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022 - Waikoloa, United States|
Duration: 2022 Jan 4 → 2022 Jan 8
|Name||Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022|
|Conference||22nd IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022|
|Period||22/1/4 → 22/1/8|
Bibliographical notePublisher Copyright:
© 2022 IEEE.
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
- Computer Science Applications