ElasticFormer：通过弹性计算视觉 Transformer 检测高分辨率宽幅图像中的目标
ElasticFormer: Detecting Objects in HRW Shots via Elastic Computing Vision Transformer
ElasticFormer 的 ElasticSelector 用方差打分加前景比例预测来动态选择窗口，这种按前景稀疏度弹性分配算力的思路可以直接迁移到任何高分辨率或大视场图像检测的骨干网络设计中，也提供了在缺少边界框标注时训练选择器的三阶段策略。
原文依据 we introduce ElasticFormer, a sparse backbone that dynamically allocates computational resources based on foreground proportion
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期十亿像素级成像的进展使高分辨率宽幅图像成为研究前沿。然而，这类图像带来了显著挑战：前景极度稀疏、十亿像素级分辨率以及目标数量多样。这使得传统的近景检测器不准确且缓慢，因为它们被背景所淹没。尽管已有研究探索了稀疏骨干网络，但其固定的稀疏模式缺乏应对多样目标数量所需的适应性。为此，我们提出 ElasticFormer，一种根据前景比例动态分配计算资源的稀疏骨干网络。在基于方差对窗口打分后，所提出的 ElasticSelector 模块会预测前景比例以进行 top-k 选择。该机制引导模型选择包含目标的窗口，在目标聚集的区域扩展资源。我们引入一种新的损失函数并结合三阶段训练策略用于 ElasticSelector，使其在缺少边界框标注时也能正常工作。我们在 PASCAL VOC 2007 上开展了一项弱监督目标检测研究以评估其可扩展性。此外，我们构建了 ElasticNet 以验证其骨干网络无关性。在 PANDA 十亿像素基准上的实验中，与固定比例稀疏方法相比，ElasticFormer 将骨干网络浮点运算量降低 80%，同时实现 AP_50 的显著提升。
英文原文摘要 1425 字符
Recent advances in gigapixel-level imaging have brought High-Resolution Wide shots to the forefront of research. However, these images present significant challenges: extreme sparsity of foreground, gigapixel-level resolutions and diverse target counts. This makes traditional close-up detectors inaccurate and slow as they are overwhelmed by the background. Although previous research has explored sparse backbones, their fixed sparsity patterns lack the adaptability required to handle diverse target numbers. To address this, we introduce ElasticFormer, a sparse backbone that dynamically allocates computational resources based on foreground proportion. After scoring windows based on variance, proposed ElasticSelector module will predict the foreground proportion for top-k selection. The mechanism guides the model to select target-containing windows, scaling resources in areas where objects are clustered. We introduce a novel loss function combined with the 3-phase training strategy for ElasticSelector, allowing it to function properly when bounding box annotations are missing. A WSOD study is carried on PASCAL VOC 2007 to evaluate its extensibility. Further, ElasticNet is created to verify its backbone-agnostic nature. In experiments on the PANDA gigapixel benchmark, ElasticFormer reduces backbone FLOPs by 80% while achieving a significant improvement in AP_ 50 when compared to fixed-ratio sparse methods.