Geometry Constrained Weakly Supervised Object Localization
This work addresses object localization with weak supervision, which is a key problem in computer vision for reducing annotation costs, but it appears incremental as it builds on existing WSOL methods with geometric constraints.
The authors tackled weakly supervised object localization by proposing GC-Net, a geometry-constrained network that predicts object locations using geometric shapes and a multi-task loss, achieving state-of-the-art performance on CUB-200-2011 and ILSVRC2012 datasets with significant margins.
We propose a geometry constrained network, termed GC-Net, for weakly supervised object localization (WSOL). GC-Net consists of three modules: a detector, a generator and a classifier. The detector predicts the object location defined by a set of coefficients describing a geometric shape (i.e. ellipse or rectangle), which is geometrically constrained by the mask produced by the generator. The classifier takes the resulting masked images as input and performs two complementary classification tasks for the object and background. To make the mask more compact and more complete, we propose a novel multi-task loss function that takes into account area of the geometric shape, the categorical cross-entropy and the negative entropy. In contrast to previous approaches, GC-Net is trained end-to-end and predict object location without any post-processing (e.g. thresholding) that may require additional tuning. Extensive experiments on the CUB-200-2011 and ILSVRC2012 datasets show that GC-Net outperforms state-of-the-art methods by a large margin. Our source code is available at https://github.com/lwzeng/GC-Net.