Object-centric architectures can learn to extract distinct object representations from visual scenes, enabling downstream applications on the object level. Similarly to autoencoder-based image models, object-centric approaches have been trained on the unsupervised reconstruction loss of images encoded by RGB color spaces. In our work, we challenge the common assumption that RGB images are the optimal color space for unsupervised learning in computer vision. We discuss conceptually and empirically that other color spaces, such as HSV, bear essential characteristics for object-centric representation learning, like robustness to lighting conditions. We further show that models improve when requiring them to predict additional color channels. Specifically, we propose to transform the predicted targets to the RGB-S space, which extends RGB with HSV's saturation component and leads to markedly better reconstruction and disentanglement for five common evaluation datasets. The use of composite color spaces can be implemented with basically no computational overhead, is agnostic of the models' architecture, and is universally applicable across a wide range of visual computing tasks and training types. The findings of our approach encourage additional investigations in computer vision tasks beyond object-centric learning.

翻译：目标中心架构能够学习从视觉场景中提取独立的目标表征，从而支持目标级别的下游应用。与基于自编码器的图像模型类似，目标中心方法通常通过RGB色彩空间编码的图像在无监督重建损失上进行训练。本研究对"RGB图像是计算机视觉无监督学习最优色彩空间"这一普遍假设提出挑战。我们从概念和实证角度论证了其他色彩空间（如HSV）具备目标中心表征学习所需的关键特性，例如对光照条件的鲁棒性。进一步研究表明，当要求模型预测额外颜色通道时，其性能会得到提升。具体而言，我们提出将预测目标转换至RGB-S空间——该空间通过融合HSV的饱和度分量扩展了RGB色彩空间，在五个常用评估数据集上实现了显著改善的重建效果和解耦性能。复合色彩空间的应用几乎不产生计算开销，与模型架构无关，且可广泛适用于各类视觉计算任务和训练范式。本方法的发现为超越目标中心学习的计算机视觉任务研究提供了新的探索方向。