Zero-shot learning (ZSL) tackles the unseen class recognition problem, transferring semantic knowledge from seen classes to unseen ones. Typically, to guarantee desirable knowledge transfer, a common (latent) space is adopted for associating the visual and semantic domains in ZSL. However, existing common space learning methods align the semantic and visual domains by merely mitigating distribution disagreement through one-step adaptation. This strategy is usually ineffective due to the heterogeneous nature of the feature representations in the two domains, which intrinsically contain both distribution and structure variations. To address this and advance ZSL, we propose a novel hierarchical semantic-visual adaptation (HSVA) framework. Specifically, HSVA aligns the semantic and visual domains by adopting a hierarchical two-step adaptation, i.e., structure adaptation and distribution adaptation. In the structure adaptation step, we take two task-specific encoders to encode the source data (visual domain) and the target data (semantic domain) into a structure-aligned common space. To this end, a supervised adversarial discrepancy (SAD) module is proposed to adversarially minimize the discrepancy between the predictions of two task-specific classifiers, thus making the visual and semantic feature manifolds more closely aligned. In the distribution adaptation step, we directly minimize the Wasserstein distance between the latent multivariate Gaussian distributions to align the visual and semantic distributions using a common encoder. Finally, the structure and distribution adaptation are derived in a unified framework under two partially-aligned variational autoencoders. Extensive experiments on four benchmark datasets demonstrate that HSVA achieves superior performance on both conventional and generalized ZSL. The code is available at \url{https://github.com/shiming-chen/HSVA} .
翻译:零点学习 (ZSL) 解决了隐蔽的阶级识别问题, 将语义学知识从可见的阶级转移到看不见的阶级。 通常, 为了保证理想的知识转移, 通常会采用一个普通( 后期) 空间空间来将 ZSL 的视觉和语义域联系起来。 但是, 现有的共同空间学习方法将语义和视觉域统一起来, 仅仅通过一步适应来缓解分布分歧。 由于两个域的地貌表达形式性质各异, 本质上包含分布和结构上的一般空间。 为了解决这个问题和推进 ZSL, 我们提出了一个新的等级的语义- 视觉适应( HSVA) 框架。 具体而言, HSVVA 将语义学和视觉- 视觉- 时间值分配框架对等起来。 在结构调整步骤中, 我们用两个特定的任务编码将源数据( 视觉域) 和目标数据( 数学域域域) 输入一个结构- 与结构一致的通用空间。 对于这个目的, 监督的对立差异( SAD) 模块在对立性 的, 在Slevilalalalalalal 分配中, ladeal ladeal deal ladeal dal dal dal laveal laveal dal dal dal laveal dal dal laut laut laut sal laut laut laut laut laut laut laut laut laut sal laut laut laut laut laut the laut laut laut laut lad lad laut lauts lauts lauts lauts lauts laut laut laut laut lauts lauts laut laut laut laut lauts lauts laut laut lauts lauts lauts laut laut laut la la la la la la lad la lad laut la la la la la la la la la la