Multi-Label Guided Soft Contrastive Learning for Efficient Earth Observation Pretraining

• URL: http://arxiv.org/abs/2405.20462v2
• Date: Mon, 23 Sep 2024 18:36:33 GMT
• Title: Multi-Label Guided Soft Contrastive Learning for Efficient Earth Observation Pretraining
• Authors: Yi Wang, Conrad M Albrecht, Xiao Xiang Zhu,
• Abstract summary: Land-cover-land-use products provide free global semantic information, as well as vision foundation models that hold strong knowledge of the natural world. We show these free additional resources not only help resolve common contrastive learning bottlenecks, but also significantly boost the efficiency and effectiveness of EO pretraining. We produce multispectral and SAR foundation models that achieve significantly better results in 10 out of 11 downstream tasks than most existing SOTA models.
• Score: 19.143105229950976
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-supervised pretraining on large-scale satellite data has raised great interest in building Earth observation (EO) foundation models. However, many important resources beyond pure satellite imagery, such as land-cover-land-use products that provide free global semantic information, as well as vision foundation models that hold strong knowledge of the natural world, are not widely studied. In this work, we show these free additional resources not only help resolve common contrastive learning bottlenecks, but also significantly boost the efficiency and effectiveness of EO pretraining. Specifically, we first propose soft contrastive learning that optimizes cross-scene soft similarity based on land-cover-generated multi-label supervision, naturally solving the issue of multiple positive samples and too strict positive matching in complex scenes. Second, we revisit and explore cross-domain continual pretraining for both multispectral and SAR imagery, building efficient EO foundation models from strongest vision models such as DINOv2. Adapting simple weight-initialization and Siamese masking strategies into our soft contrastive learning framework, we demonstrate impressive continual pretraining performance even when the input modalities are not aligned. Without prohibitive training, we produce multispectral and SAR foundation models that achieve significantly better results in 10 out of 11 downstream tasks than most existing SOTA models. For example, our ResNet50/ViT-S achieve 84.8/85.0 linear probing mAP scores on BigEarthNet-10\% which are better than most existing ViT-L models; under the same setting, our ViT-B sets a new record of 86.8 in multispectral, and 82.5 in SAR, the latter even better than many multispectral models. Dataset and models are available at \url{https://github.com/zhu-xlab/softcon}.

Related papers

• Efficient Self-Supervised Learning for Earth Observation via Dynamic Dataset Curation [67.23953699167274]
  Self-supervised learning (SSL) has enabled the development of vision foundation models for Earth Observation (EO) In EO, this challenge is amplified by the redundancy and heavy-tailed distributions common in satellite imagery. We propose a dynamic dataset pruning strategy designed to improve SSL pre-training by maximizing dataset diversity and balance.
  arXiv  Detail & Related papers  (2025-04-09T15:13:26Z)
• SatVision-TOA: A Geospatial Foundation Model for Coarse-Resolution All-Sky Remote Sensing Imagery [8.096413986108601]
  We introduce SatVision-TOA, a novel foundation model pre-trained on 14-band MODIS L1B Top-Of-Atmosphere (TOA) radiance imagery. The SatVision-TOA model is pre-trained using a Masked-Image-Modeling (MIM) framework and the SwinV2 architecture. Results show that SatVision-TOA achieves superior performance over baseline methods on downstream tasks.
  arXiv  Detail & Related papers  (2024-11-26T00:08:00Z)
• An Experimental Study on Exploring Strong Lightweight Vision Transformers via Masked Image Modeling Pre-Training [51.622652121580394]
  Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the textitextremely simple lightweight ViTs' fine-tuning performance can also benefit from this pre-training paradigm. Our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design ($5.7M$/$6.5M$) can achieve $79.4%$/$78.9%$ top-1 accuracy on ImageNet-1
  arXiv  Detail & Related papers  (2024-04-18T14:14:44Z)
• Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery [78.43828998065071]
  Recent advances in unsupervised learning have demonstrated the ability of large vision models to achieve promising results on downstream tasks. Such pre-training techniques have also been explored recently in the remote sensing domain due to the availability of large amount of unlabelled data. In this paper, we re-visit transformers pre-training and leverage multi-scale information that is effectively utilized with multiple modalities.
  arXiv  Detail & Related papers  (2024-03-08T16:18:04Z)
• MoE-LLaVA: Mixture of Experts for Large Vision-Language Models [27.930351465266515]
  We propose a simple yet effective training strategy MoE-Tuning for LVLMs. MoE-LLaVA, a MoE-based sparse LVLM architecture, uniquely activates only the top-k experts through routers. Experiments show the significant performance of MoE-LLaVA in a variety of visual understanding and object hallucination benchmarks.
  arXiv  Detail & Related papers  (2024-01-29T08:13:40Z)
• Optimization Efficient Open-World Visual Region Recognition [55.76437190434433]
  RegionSpot integrates position-aware localization knowledge from a localization foundation model with semantic information from a ViL model. Experiments in open-world object recognition show that our RegionSpot achieves significant performance gain over prior alternatives.
  arXiv  Detail & Related papers  (2023-11-02T16:31:49Z)
• Foundation Models for Generalist Geospatial Artificial Intelligence [3.7002058945990415]
  This paper introduces a first-of-a-kind framework for the efficient pre-training and fine-tuning of foundational models on extensive data. We have utilized this framework to create Prithvi, a transformer-based foundational model pre-trained on more than 1TB of multispectral satellite imagery.
  arXiv  Detail & Related papers  (2023-10-28T10:19:55Z)
• The effectiveness of MAE pre-pretraining for billion-scale pretraining [65.98338857597935]
  We introduce an additional pre-pretraining stage that is simple and uses the self-supervised MAE technique to initialize the model. We measure the effectiveness of pre-pretraining on 10 different visual recognition tasks spanning image classification, video recognition, object detection, low-shot classification and zero-shot recognition.
  arXiv  Detail & Related papers  (2023-03-23T17:56:12Z)
• eP-ALM: Efficient Perceptual Augmentation of Language Models [70.47962271121389]
  We propose to direct effort to efficient adaptations of existing models, and propose to augment Language Models with perception. Existing approaches for adapting pretrained models for vision-language tasks still rely on several key components that hinder their efficiency. We show that by freezing more than 99% of total parameters, training only one linear projection layer, and prepending only one trainable token, our approach (dubbed eP-ALM) significantly outperforms other baselines on VQA and Captioning.
  arXiv  Detail & Related papers  (2023-03-20T19:20:34Z)
• Virtual embeddings and self-consistency for self-supervised learning [43.086696088061416]
  TriMix is a novel concept for self-supervised learning that generates virtual embeddings through linear data. We validate TriMix on eight benchmark datasets with an improvement of 2.71% and 0.41% better than the second-best models for both data types.
  arXiv  Detail & Related papers  (2022-06-13T10:20:28Z)
• Sparse MoEs meet Efficient Ensembles [49.313497379189315]
  We study the interplay of two popular classes of such models: ensembles of neural networks and sparse mixture of experts (sparse MoEs) We present Efficient Ensemble of Experts (E$3$), a scalable and simple ensemble of sparse MoEs that takes the best of both classes of models, while using up to 45% fewer FLOPs than a deep ensemble.
  arXiv  Detail & Related papers  (2021-10-07T11:58:35Z)

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.