Cycle-Object Consistency for Image-to-Image Domain Adaptation
Artikel i vetenskaplig tidskrift, 2023

Recent advances in generative adversarial networks (GANs) have been proven effective in performing domain adaptation for object detectors through data augmentation. While GANs are exceptionally successful, those methods that can preserve objects well in the image-to-image translation task usually require an auxiliary task, such as semantic segmentation to prevent the image content from being too distorted. However, pixel-level annotations are difficult to obtain in practice. Alternatively, instance-aware image-translation model treats object instances and background separately. Yet, it requires object detectors at test time, assuming that off-the-shelf detectors work well in both domains. In this work, we present AugGAN-Det, which introduces Cycle-object Consistency (CoCo) loss to generate instance-aware translated images across complex domains. The object detector of the target domain is directly leveraged in generator training and guides the preserved objects in the translated images to carry target-domain appearances. Compared to previous models, which e.g., require pixel-level semantic segmentation to force the latent distribution to be object-preserving, this work only needs bounding box annotations which are significantly easier to acquire. Next, as to the instance-aware GAN models, our model, AugGAN-Det, internalizes global and object style-transfer without explicitly aligning the instance features. Most importantly, a detector is not required at test time. Experimental results demonstrate that our model outperforms recent object-preserving and instance-level models and achieves state-of-the-art detection accuracy and visual perceptual quality.

Cross-domain object detection

Instance-aware image-translation

Domain adaptation

Generative adversarial network

Författare

Che-Tsung Lin

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik

Jie Long Kew

Chee Seng Chan

Shang Hong Lai

Christopher Zach

Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik

Pattern Recognition

0031-3203 (ISSN)

Vol. 138 109416

Ämneskategorier

Människa-datorinteraktion (interaktionsdesign)

Elektroteknik och elektronik

Signalbehandling

Diskret matematik

DOI

10.1016/j.patcog.2023.109416

Mer information

Senast uppdaterat

2023-04-12