Object Learning and Robust 3D Reconstruction
This work addresses the problem of unsupervised object learning and robust 3D reconstruction for computer vision researchers, presenting incremental advancements in object-based approaches.
The thesis tackles unsupervised object segmentation in 2D using motion cues and extends to 3D reconstruction by detecting and removing dynamic objects to improve geometric consistency, achieving robust 3D modeling in casual capture setups.
In this thesis we discuss architectural designs and training methods for a neural network to have the ability of dissecting an image into objects of interest without supervision. The main challenge in 2D unsupervised object segmentation is distinguishing between foreground objects of interest and background. FlowCapsules uses motion as a cue for the objects of interest in 2D scenarios. The last part of this thesis focuses on 3D applications where the goal is detecting and removal of the object of interest from the input images. In these tasks, we leverage the geometric consistency of scenes in 3D to detect the inconsistent dynamic objects. Our transient object masks are then used for designing robust optimization kernels to improve 3D modelling in a casual capture setup. One of our goals in this thesis is to show the merits of unsupervised object based approaches in computer vision. Furthermore, we suggest possible directions for defining objects of interest or foreground objects without requiring supervision. Our hope is to motivate and excite the community into further exploring explicit object representations in image understanding tasks.