On the Viability of Monocular Depth Pre-training for Semantic Segmentation

• URL: http://arxiv.org/abs/2203.13987v5
• Date: Thu, 18 Jul 2024 05:36:22 GMT
• Title: On the Viability of Monocular Depth Pre-training for Semantic Segmentation
• Authors: Dong Lao, Fengyu Yang, Daniel Wang, Hyoungseob Park, Samuel Lu, Alex Wong, Stefano Soatto,
• Abstract summary: We study whether pre-training on geometric tasks is viable for downstream transfer to semantic tasks. We find that monocular depth is a viable form of pre-training for semantic segmentation, validated by improvements over common baselines.
• Score: 48.29060171161375
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The question of whether pre-training on geometric tasks is viable for downstream transfer to semantic tasks is important for two reasons, one practical and the other scientific. If the answer is positive, we may be able to reduce pre-training cost and bias from human annotators significantly. If the answer is negative, it may shed light on the role of embodiment in the emergence of language and other cognitive functions in evolutionary history. To frame the question in a way that is testable with current means, we pre-train a model on a geometric task, and test whether that can be used to prime a notion of 'object' that enables inference of semantics as soon as symbols (labels) are assigned. We choose monocular depth prediction as the geometric task, and semantic segmentation as the downstream semantic task, and design a collection of empirical tests by exploring different forms of supervision, training pipelines, and data sources for both depth pre-training and semantic fine-tuning. We find that monocular depth is a viable form of pre-training for semantic segmentation, validated by improvements over common baselines. Based on the findings, we propose several possible mechanisms behind the improvements, including their relation to dataset size, resolution, architecture, in/out-of-domain source data, and validate them through a wide range of ablation studies. We also find that optical flow, which at first glance may seem as good as depth prediction since it optimizes the same photometric reprojection error, is considerably less effective, as it does not explicitly aim to infer the latent structure of the scene, but rather the raw phenomenology of temporally adjacent images.

Related papers

• Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps [39.00415825387414]
  We propose a new approach for semantic correspondence estimation that supplements discriminative features with 3D understanding via a weak geometric spherical prior. Compared to more involved 3D pipelines, our model only requires weak viewpoint information, and the simplicity of our spherical representation enables us to inject informative geometric priors into the model during training. We present results on the challenging SPair-71k dataset, where our approach demonstrates is capable of distinguishing between symmetric views and repeated parts across many object categories.
  arXiv  Detail & Related papers  (2023-12-20T17:35:24Z)
• Inverse Dynamics Pretraining Learns Good Representations for Multitask Imitation [66.86987509942607]
  We evaluate how such a paradigm should be done in imitation learning. We consider a setting where the pretraining corpus consists of multitask demonstrations. We argue that inverse dynamics modeling is well-suited to this setting.
  arXiv  Detail & Related papers  (2023-05-26T14:40:46Z)
• Self-Supervised Learning via Maximum Entropy Coding [57.56570417545023]
  We propose Maximum Entropy Coding (MEC) as a principled objective that explicitly optimize on the structure of the representation. MEC learns a more generalizable representation than previous methods based on specific pretext tasks. It achieves state-of-the-art performance consistently on various downstream tasks, including not only ImageNet linear probe, but also semi-supervised classification, object detection, instance segmentation, and object tracking.
  arXiv  Detail & Related papers  (2022-10-20T17:58:30Z)
• Self-Distillation for Further Pre-training of Transformers [83.84227016847096]
  We propose self-distillation as a regularization for a further pre-training stage. We empirically validate the efficacy of self-distillation on a variety of benchmark datasets for image and text classification tasks.
  arXiv  Detail & Related papers  (2022-09-30T02:25:12Z)
• Masked prediction tasks: a parameter identifiability view [49.533046139235466]
  We focus on the widely used self-supervised learning method of predicting masked tokens. We show that there is a rich landscape of possibilities, out of which some prediction tasks yield identifiability, while others do not.
  arXiv  Detail & Related papers  (2022-02-18T17:09:32Z)
• Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision [57.14468881854616]
  We propose an auxiliary training objective that improves the generalization capabilities of neural networks. We use pairs of minimally-different examples with different labels, a.k.a counterfactual or contrasting examples, which provide a signal indicative of the underlying causal structure of the task. Models trained with this technique demonstrate improved performance on out-of-distribution test sets.
  arXiv  Detail & Related papers  (2020-04-20T02:47:49Z)
• Semantically-Guided Representation Learning for Self-Supervised Monocular Depth [40.49380547487908]
  We propose a new architecture leveraging fixed pretrained semantic segmentation networks to guide self-supervised representation learning. Our method improves upon the state of the art for self-supervised monocular depth prediction over all pixels, fine-grained details, and per semantic categories.
  arXiv  Detail & Related papers  (2020-02-27T18:40:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.