用少量参数实现高效弱监督目标定位，提升物体覆盖度。
TriLite: Efficient Weakly Supervised Object Localization with Universal Visual Features and Tri-Region Disentanglement
- 冻结预训练ViT，仅新增不足80万可训练参数
- 三区域解耦模块提升物体定位覆盖，抑制误激活
- 单阶段框架，训练简单且参数效率高，适合资源有限场景
弱监督目标定位（WSOL）仅用图像级标签定位图像中的目标物体。尽管近期取得进展，许多方法仍依赖多阶段流程或对大型骨干网络进行全微调，导致训练成本高，且普遍存在部分物体覆盖不足的问题。本文提出TriLite，一种单阶段WSOL框架，利用自监督预训练的Dinov2 Vision Transformer作为冻结主干，并仅引入极少可训练参数（ImageNet-1K上少于800K），同时完成分类与定位任务。核心是提出的TriHead模块，将图像块特征分解为前景、背景和模糊区域，从而提升物体覆盖度并抑制伪激活。通过解耦分类与定位目标，TriLite有效利用自监督ViT学习到的通用表征，无需昂贵的端到端训练。在CUB-200-2011、ImageNet-1K和OpenImages上的实验表明，TriLite达到新最优性能，同时显著更参数高效且易于训练。代码即将开源。
原文摘要 · Abstract (English)
Weakly supervised object localization (WSOL) aims to localize target objects in images using only image-level labels. Despite recent progress, many approaches still rely on multi-stage pipelines or full fine-tuning of large backbones, which increases training cost, while the broader WSOL community continues to face the challenge of partial object coverage. We present TriLite, a single-stage WSOL framework that leverages a frozen Vision Transformer with Dinov2 pre-training in a self-supervised manner, and introduces only a minimal number of trainable parameters (fewer than 800K on ImageNet-1K) for both classification and localization. At its core is the proposed TriHead module, which decomposes patch features into foreground, background, and ambiguous regions, thereby improving object coverage while suppressing spurious activations. By disentangling classification and localization objectives, TriLite effectively exploits the universal representations learned by self-supervised ViTs without requiring expensive end-to-end training. Extensive experiments on CUB-200-2011, ImageNet-1K, and OpenImages demonstrate that TriLite sets a new state of the art, while remaining significantly more parameter-efficient and easier to train than prior methods. The code will be released soon.
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。