Sep 21 2025
A new AI framework overcomes motion blur in handheld videos, pushing NeRF technology toward sharper 3D content for VR, AR, robotics, and everyday smartphone captures.
Image Credit: WhiplashRick / Shutterstock
What is NeRF?
NeRF is a fascinating technique that creates three-dimensional (3D) representations of a scene from a set of two-dimensional (2D) images, captured from different angles. It works by training a deep neural network to predict the color and density at any point in 3D space. To do this, it casts imaginary light rays from the camera through each pixel in all input images, sampling points along those rays with their 3D coordinates and viewing direction. Using this information, NeRF reconstructs the scene in 3D and can render it from entirely new perspectives, a process known as novel view synthesis (NVS).
Challenges with Blurry Videos
Beyond still images, a video can also be used, with each frame treated as a static image. However, existing methods are highly sensitive to the quality of the videos. Videos captured with a single camera, such as those from a phone or drone, often suffer from motion blur due to fast object motion or camera shake. This makes it challenging to create sharp, dynamic NVS. This is because most existing deblurring-based NVS methods are designed for static multi-view images, which fail to account for global camera and local object motion. Additionally, blurry videos often result in inaccurate camera pose estimations and a loss of geometric precision.
Introducing MoBluRF
To address these issues, a research team jointly led by Assistant Professor Jihyong Oh from the Graduate School of Advanced Imaging Science (GSIAM) at Chung-Ang University (CAU) in Korea, and Professor Munchurl Kim from Korea Advanced Institute of Science and Technology (KAIST), Korea, along with Mr. Minh-Quan Viet Bui and Mr. Jongmin Park, developed MoBluRF, a two-stage motion deblurring method for NeRFs. "Our framework is capable of reconstructing sharp 4D scenes and enabling NVS from blurry monocular videos using motion decomposition, while avoiding mask supervision, significantly advancing the NeRF field," explains Dr. Oh. Their study is published in Volume 47, Issue 09 of the IEEE Transactions on Pattern Analysis and Machine Intelligence on September 01, 2025.
Two-Stage Deblurring Framework
MoBluRF consists of two main stages: BRI and MDD. Existing deblurring-based NVS methods attempt to predict hidden sharp light rays in blurry images, known as latent sharp rays, by transforming a base ray. However, directly using input rays in blurry images as base rays can lead to inaccurate prediction. BRI addresses this issue by roughly reconstructing dynamic 3D scenes from blurry videos and refining the initialization of "base rays" from imprecise camera rays.
Next, these base rays are utilized in the MDD stage to predict latent sharp rays using an ILSP method accurately. ILSP incrementally decomposes motion blur into global camera motion and local object motion components, thereby significantly improving deblurring accuracy. MoBluRF also introduces two novel loss functions: one that separates static and dynamic regions without motion masks, and another that enhances the geometric accuracy of dynamic objects, two areas where previous methods have struggled.
Performance and Results
Owing to this innovative design, MoBluRF outperforms state-of-the-art methods with significant margins in various datasets, both quantitatively and qualitatively. It is also robust against varying degrees of blur.
Applications and Future Directions
"By enabling deblurring and 3D reconstruction from casual handheld captures, our framework enables smartphones and other consumer devices to produce sharper and more immersive content," remarks Dr. Oh. "It could also help create crisp 3D models of shaky footages from museums, improve scene understanding and safety for robots and drones, and reduce the need for specialized capture setups in virtual and augmented reality."
MoBluRF marks a new direction for NeRFs, enabling high-quality 3D reconstructions from ordinary blurry videos recorded with everyday devices.
Source:
Journal reference:
- M. -Q. V. Bui, J. Park, J. Oh and M. Kim, "MoBluRF: Motion Deblurring Neural Radiance Fields for Blurry Monocular Video," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 47, no. 9, pp. 7752-7770, Sept. 2025, doi: 10.1109/TPAMI.2025.3574644, https://ieeexplore.ieee.org/document/11017407
UT San Antonio Launches College Of AI Cyber And Computing To Drive Tech Innovation And Workforce Growth