Discriminative Region-based Multi-Label Zero-Shot Learning

TL;DR

This paper introduces a novel region-based multi-label zero-shot learning method that preserves spatial details and uses a bi-level attention module to improve class discriminability, achieving state-of-the-art results.

Contribution

It proposes a discriminability-preserving approach with region-level features and bi-level attention, addressing limitations of shared attention maps in multi-label ZSL.

Findings

Achieves 6.9% higher mAP on NUS-WIDE for ZSL.

Sets new state-of-the-art on NUS-WIDE and Open Images benchmarks.

Improves discriminability and reduces feature entanglement in multi-label ZSL.

Abstract

Multi-label zero-shot learning (ZSL) is a more realistic counter-part of standard single-label ZSL since several objects can co-exist in a natural image. However, the occurrence of multiple objects complicates the reasoning and requires region-specific processing of visual features to preserve their contextual cues. We note that the best existing multi-label ZSL method takes a shared approach towards attending to region features with a common set of attention maps for all the classes. Such shared maps lead to diffused attention, which does not discriminatively focus on relevant locations when the number of classes are large. Moreover, mapping spatially-pooled visual features to the class semantics leads to inter-class feature entanglement, thus hampering the classification. Here, we propose an alternate approach towards region-based discriminability-preserving multi-label zero-shot classification. Our approach maintains the spatial resolution to preserve region-level characteristics and utilizes a bi-level attention module (BiAM) to enrich the features by incorporating both region and scene context information. The enriched region-level features are then mapped to the class semantics and only their class predictions are spatially pooled to obtain image-level predictions, thereby keeping the multi-class features disentangled. Our approach sets a new state of the art on two large-scale multi-label zero-shot benchmarks: NUS-WIDE and Open Images. On NUS-WIDE, our approach achieves an absolute gain of 6.9% mAP for ZSL, compared to the best published results.