Restoring Real-World Degraded Events Improves Deblurring Quality

Yeqing Shen, Shang Li, Kun Song

Abstract: Due to its high speed and low latency, DVS is frequently employed in motion deblurring. Ideally, high-quality events would adeptly capture intricate motion information. However, real-world events are generally degraded, thereby introducing significant artifacts into the deblurred results. In response to this challenge, we model the degradation of events and propose RDNet to improve the quality of image deblurring. Specifically, we first analyze the mechanisms underlying degradation and simulate paired events based on that. These paired events are then fed into the first stage of the RDNet for training the restoration model. The events restored in this stage serve as a guide for the second-stage deblurring process. To better assess the deblurring performance of different methods on real-world degraded events, we present a new real-world dataset named DavisMCR. This dataset incorporates events with diverse degradation levels, collected by manipulating environmental brightness and target object contrast. Our experiments are conducted on synthetic datasets (GOPRO), real-world datasets (REBlur), and the proposed dataset (DavisMCR). The results demonstrate that RDNet outperforms classical event denoising methods in event restoration. Furthermore, RDNet exhibits better performance in deblurring tasks compared to state-of-the-art methods. DavisMCR are available at https://github.com/Yeeesir/DVS_RDNet.

Primary Subject Area: [Content] Media Interpretation

Secondary Subject Area: [Content] Multimodal Fusion

Relevance To Conference:

Image processing and event camera-based image reconstruction are crucial components in multimedia and multimodal processing. Traditional image processing techniques are adept at manipulating static images captured by conventional cameras, facilitating tasks such as enhancement, denoising, and object recognition. However, these methods encounter limitations when dealing with dynamic scenes or fast-moving objects due to motion blur and high temporal resolution requirements.
Event cameras offer a transformative solution by capturing visual information asynchronously at the pixel level, driven by changes in brightness (events), instead of capturing entire frames at fixed intervals. This asynchronous nature enables event cameras to excel in high-speed environments, low-light conditions, and scenes with rapid changes, where traditional cameras struggle. Event-based imaging also consumes significantly less power and bandwidth compared to conventional cameras, making them ideal for resource-constrained applications.
By integrating image processing techniques with event camera-based image reconstruction, multimedia and multimodal processing systems can achieve unprecedented levels of performance and versatility. These technologies enable applications such as real-time object tracking, gesture recognition, augmented reality, and autonomous navigation in dynamic environments. Moreover, they hold promise for emerging fields like healthcare, robotics, and surveillance, where accurate and efficient processing of visual information is paramount. As research in this area continues to evolve, the integration of image processing and event-based imaging will further revolutionize multimedia and multimodal processing, opening new avenues for innovation and discovery.

Submission Number: 1633