用简单蒸馏法让模型生成更高分辨率的密集特征，降低成本。
Another BRIXEL in the Wall: Towards Cheaper Dense Features
- 学生模型通过自蒸馏学习在更高分辨率下复现特征图。
- 固定分辨率下，性能显著超越DINOv3基线模型。
- 适用于多种密集特征提取器，提升广泛适用性。
视觉基础模型在全局和局部密集下游任务中表现优异。近期的DINOv3系列模型基于大量图像预训练，可生成极为精细的密集特征图，实现顶尖性能。然而，生成这些特征图需高分辨率输入，并因Transformer架构的平方复杂度导致计算量巨大。为此，我们提出BRIXEL——一种简单的知识蒸馏方法，使学生模型学习在更高分辨率下重现自身特征图。尽管方法简单，但在分辨率固定时，BRIXEL在下游任务中大幅超越基线DINOv3模型。我们还将BRIXEL应用于其他近期密集特征提取器，结果表明其在多个模型家族中均带来显著性能提升。代码与模型权重见https://github.com/alexanderlappe/BRIXEL。
原文摘要 · Abstract (English)
Vision foundation models achieve strong performance on both global and locally dense downstream tasks. Pretrained on large images, the recent DINOv3 model family is able to produce very fine-grained dense feature maps, enabling state-of-the-art performance. However, computing these feature maps requires the input image to be available at very high resolution, as well as large amounts of compute due to the squared complexity of the transformer architecture. To address these issues, we propose BRIXEL, a simple knowledge distillation approach that has the student learn to reproduce its own feature maps at higher resolution. Despite its simplicity, BRIXEL outperforms the baseline DINOv3 models by large margins on downstream tasks when the resolution is kept fixed. We also apply BRIXEL to other recent dense-feature extractors and show that it yields substantial performance gains across model families. Code and model weights are available at https://github.com/alexanderlappe/BRIXEL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。