# Composite Learning for Robust and Effective Dense Predictions Source: [Composite Learning for Robust and Effective Dense Predictions]() Collection index: [llms.txt]() Data updated: 2026-01-04T03:47:11.955Z Authors: Menelaos Kanakis, Thomas E. Huang, David Bruggemann, Fisher Yu, Luc Van Gool Published: 2022-10-13T17:59:16.000Z ## Abstract Multi-task learning promises better model generalization on a target task by jointly optimizing it with an auxiliary task. However, the current practice requires additional labeling efforts for the auxiliary task, while not guaranteeing better model performance. In this paper, we find that jointly training a dense prediction (target) task with a self-supervised (auxiliary) task can consistently improve the performance of the target task, while eliminating the need for labeling auxiliary tasks. We refer to this joint training as Composite Learning (CompL). Experiments of CompL on monocular depth estimation, semantic segmentation, and boundary detection show consistent performance improvements in fully and partially labeled datasets. Further analysis on depth estimation reveals that joint training with self-supervision outperforms most labeled auxiliary tasks. We also find that CompL can improve model robustness when the models are evaluated in new domains. These results demonstrate the benefits of self-supervision as an auxiliary task, and establish the design of novel task-specific self-supervised methods as a new axis of investigation for future multi-task learning research. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Semantic Segmentation on PASCAL VOC augmented 2012 (1% labeled)](): Semantic segmentation; PASCAL VOC augmented 2012 (1% labeled); mIoU: 41.41. - [Semantic Segmentation on PASCAL VOC augmented 2012 (100% labeled)](): Semantic segmentation; PASCAL VOC augmented 2012 (100% labeled); mIoU: 76.5. - [Boundary Detection on BSDS 500 (test)](): Boundary Detection; BSDS 500 (test); ODS: 76.9. - [Monocular Depth Estimation on NYU v2 (test)](): Monocular Depth Estimation; NYU v2 (test); Abs Rel: not recorded. ## Original sources and implementation [Original paper]() [PDF]()