SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.14166 · CVPR · 2021

Indoor Panorama Planar 3D Reconstruction via Divide and Conquer

Min Sun, Ning-Hsu Wang, Chi-Wei Hsiao, Hwann-Tzong Chen, Cheng Sun

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 5 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
sunset1995/panoplane360 — 5 of 7
FunctionStatusWhere it lives
Bin_Mean_Shift Ran sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("7c25ac9747c0fb7c")
FPN_refine Ran sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("ce5b6bddb078776e")
find_peaks Ran sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("33c8f55017005e9d")
radius_loss Ran sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("a4213b01aa248670")
segement_by_rad Ran sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("d510afcba3002ede")
Net Not yet run sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("77a4b881d28a3400")
hinge_embedding_loss Not yet run sunset1995/panoplane360/models/pano_plane_360.py
pointer only (licence: NONE) · get_code("a21a95354281c4f1")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Indoor panorama typically consists of human-made structures parallel or perpendicular to gravity. We leverage this phenomenon to approximate the scene in a 360-degree image with (H)orizontal-planes and (V)ertical-planes. To this end, we propose an effective divide-and-conquer strategy that divides pixels based on their plane orientation estimation; then, the succeeding instance segmentation module conquers the task of planes clustering more easily in each plane orientation group. Besides, parameters of V-planes depend on camera yaw rotation, but translation-invariant CNNs are less aware of the yaw change. We thus propose a yaw-invariant V-planar reparameterization for CNNs to learn. We create a benchmark for indoor panorama planar reconstruction by extending existing 360 depth datasets with ground truth H&V-planes (referred to as "PanoH&V" dataset) and adopt state-of-the-art planar reconstruction methods to predict H&V-planes as our baselines. Our method outperforms the baselines by a large margin on the proposed dataset.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2106.14166")
get_code_for_paper("2106.14166")
have("2106.14166")

Connect an agent — have() is free.