Prompt-based test-time real image dehazing: a novel pipeline
Zixuan Chen, Zewei He, Ziqian Lu, Xuecheng Sun, Zhe-Ming Lu

TL;DR
This paper introduces a novel test-time pipeline called Prompt-based Test-Time Dehazing (PTTD) that enhances real-world image dehazing performance by adjusting feature statistics during inference, without complex training.
Contribution
The paper proposes a model-agnostic, test-time dehazing pipeline using prompt generation and feature adaptation to improve real image dehazing from synthetic-trained models.
Findings
PTTD outperforms state-of-the-art methods in real-world scenarios.
Feature statistic adjustment effectively narrows the domain gap.
The approach is compatible with various dehazing models.
Abstract
Existing methods attempt to improve models' generalization ability on real-world hazy images by exploring well-designed training schemes (\eg, CycleGAN, prior loss). However, most of them need very complicated training procedures to achieve satisfactory results. For the first time, we present a novel pipeline called Prompt-based Test-Time Dehazing (PTTD) to help generate visually pleasing results of real-captured hazy images during the inference phase. We experimentally observe that given a dehazing model trained on synthetic data, fine-tuning the statistics (\ie, mean and standard deviation) of encoding features is able to narrow the domain gap, boosting the performance of real image dehazing. Accordingly, we first apply a prompt generation module (PGM) to generate a visual prompt, which is the reference of appropriate statistical perturbations for mean and standard deviation. Then, we…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsImage Enhancement Techniques · Advanced Neural Network Applications · Generative Adversarial Networks and Image Synthesis
MethodsHuMan(Expedia)||How do I get a human at Expedia? · *Communicated@Fast*How Do I Communicate to Expedia? · Residual Connection · Batch Normalization · Tanh Activation · PatchGAN · Sigmoid Activation · GAN Least Squares Loss · Residual Block · Cycle Consistency Loss
