I$^2$-SLAM: Inverting Imaging Process for Robust Photorealistic Dense SLAM

TL;DR

This paper introduces I$^2$-SLAM, an inverse imaging module that enhances visual SLAM robustness in casual scenarios by modeling physical image formation, leading to improved trajectory accuracy and photorealistic 3D reconstructions.

Contribution

It integrates physical imaging models into SLAM, jointly optimizing imaging parameters to handle motion blur and appearance changes, which is a novel approach for robust dense SLAM.

Findings

Improves SLAM robustness in casual videos with motion blur.

Produces higher quality, photorealistic 3D reconstructions.

Enhances trajectory accuracy across various scene representations.

Abstract

We present an inverse image-formation module that can enhance the robustness of existing visual SLAM pipelines for casually captured scenarios. Casual video captures often suffer from motion blur and varying appearances, which degrade the final quality of coherent 3D visual representation. We propose integrating the physical imaging into the SLAM system, which employs linear HDR radiance maps to collect measurements. Specifically, individual frames aggregate images of multiple poses along the camera trajectory to explain prevalent motion blur in hand-held videos. Additionally, we accommodate per-frame appearance variation by dedicating explicit variables for image formation steps, namely white balance, exposure time, and camera response function. Through joint optimization of additional variables, the SLAM pipeline produces high-quality images with more accurate trajectories. Extensive experiments demonstrate that our approach can be incorporated into recent visual SLAM pipelines using various scene representations, such as neural radiance fields or Gaussian splatting.