Abstract
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.
Abstract (translated)
弱监督目标定位（WSOL）旨在仅使用图像级标签来识别图像中的目标对象。尽管近年来取得了一些进展，但许多方法仍然依赖于多阶段流程或大型骨干网络的完全微调，这增加了训练成本，而更广泛的WSOL社区仍在面对部分目标覆盖范围的问题挑战。 我们提出了一种名为TriLite的一阶段WSOL框架，该框架利用冻结的Vision Transformer（ViT）在自监督模式下进行Dinov2预训练，并且引入了非常少量可训练参数（少于800K个参数用于ImageNet-1K），这些参数同时支持分类和定位任务。TriLite的核心是所提出的TriHead模块，该模块将图像块特征分解为前景、背景及模棱两可区域，从而提高目标覆盖范围并抑制不必要激活。 通过分离分类和定位的目标，TriLite可以有效地利用自监督ViT学习到的通用表示，而无需昂贵的端到端训练。在CUB-200-2011、ImageNet-1K 和 OpenImages 上进行的广泛实验表明，TriLite在保持比先前方法更高效且更容易训练的同时，确立了新的技术水平。代码不久将公开发布。
URL
https://arxiv.org/abs/2602.23120
