InstaFormer: Instance-Aware Image-to-Image Translation with Transformer

TL;DR

InstaFormer introduces a Transformer-based architecture for instance-aware image-to-image translation, integrating global and object-level features with novel modules and loss functions to improve translation quality.

Contribution

The paper proposes a new Transformer architecture with instance-level feature integration and contrastive loss for enhanced image translation.

Findings

Outperforms recent methods in instance-aware image translation

Effective global and instance-level feature integration demonstrated

Ablation studies confirm the contributions of key modules

Abstract

We present a novel Transformer-based network architecture for instance-aware image-to-image translation, dubbed InstaFormer, to effectively integrate global- and instance-level information. By considering extracted content features from an image as tokens, our networks discover global consensus of content features by considering context information through a self-attention module in Transformers. By augmenting such tokens with an instance-level feature extracted from the content feature with respect to bounding box information, our framework is capable of learning an interaction between object instances and the global image, thus boosting the instance-awareness. We replace layer normalization (LayerNorm) in standard Transformers with adaptive instance normalization (AdaIN) to enable a multi-modal translation with style codes. In addition, to improve the instance-awareness and translation quality at object regions, we present an instance-level content contrastive loss defined between input and translated image. We conduct experiments to demonstrate the effectiveness of our InstaFormer over the latest methods and provide extensive ablation studies.