Decoupling Spherical Reasoning from Dense Prediction for 360 Depth Estimation

Zhijie Shen; Chunyu Lin; Shuai Zheng; Feng Li; Runmin Cong; Huihui Bai; Yao Zhao

cs.CVSep 30, 2026

Decoupling Spherical Reasoning from Dense Prediction for 360 Depth Estimation

Authors: Zhijie Shen, Chunyu Lin, Shuai Zheng, Feng Li, Runmin Cong, Huihui Bai, Yao Zhao

Organizations: Beijing Jiaotong University · Hefei University of Technology · Shandong University

Abstract

The equirectangular projection (ERP) is widely used for panoramic depth estimation, but its spatially varying distortion makes geometry-consistent feature modeling challenging. We revisit panoramic depth estimation by decoupling contextual modeling in native spherical space from dense ERP prediction. To this end, we propose a Fibonacci Spherical Graph (FSG) as an intermediate reasoning space to lift ERP features onto quasi-uniform Fibonacci nodes on the sphere and capture local and long-range dependencies through complementary spherical neighborhoods. The resulting spherical discretization distributes graph nodes approximately uniformly over the spherical surface, reducing the over-representation of highly stretched regions during relational modeling. Operating on a compact set of Fibonacci nodes also avoids the computational burden of constructing and processing a graph at full ERP resolution. To bridge spherical reasoning and dense prediction, we propose a Spherical Context Conditioning (SCC) module that adaptively modulates dense ERP features with the enhanced spherical representation, allowing spherical context to guide pixel-aligned depth prediction. Extensive experiments on three benchmarks demonstrate that the proposed method consistently achieves superior depth accuracy over existing approaches.

Figures & tables

Figure 1: Overview of the proposed framework. (a) FSG and SCC refine multi-scale ERP features before ERP–cubemap fusion. (b) FSG samples features onto Fibonacci nodes, aggregates local, dilated, and antipodal context, and back-projects the enhanced features to ERP. (c) SCC uses spherical context $G$ to condition dense ERP features $X$ through gated modulation and context injection. The schematic is elaborated by the operators and residual update in Sec. 3.4 .

Dataset	Method	Abs Rel $\downarrow$	Sq Rel $\downarrow$	RMSE $\downarrow$	$\delta_{1}(\%)$ $\uparrow$	$\delta_{2}(\%)$ $\uparrow$	$\delta_{3}(\%)$ $\uparrow$
M3D	EGFormer	0.1473	0.1517	0.6025	81.58	93.90	97.35
	PanoFormer	0.1051	0.0966	0.4929	89.08	96.23	98.31
	BiFuse	0.1126	0.0992	0.5027	88.00	96.13	98.47
	BiFuse++	0.1123	0.0915	0.4853	88.12	96.56	98.69
	UniFuse	0.1144	0.0936	0.4835	87.85	96.59	98.73
	OmniFusion	0.1161	0.1007	0.4931	87.72	96.15	98.44

Table 1: Quantitative comparison with existing panoramic depth estimation methods. For fair comparison, ∗ denotes re-evaluation under the Elite360D protocol, while † denotes retraining with a ResNet-34 backbone. All remaining baseline results are taken from PGFuse ( Shen et al., 2026 ) . Best and second-best results are shown in bold and underlined, respectively.

Figure 2: Qualitative comparison results on Matterport3D, Stanford2D3D, and Structured3D. Best viewed in color.

Support	Graph	SCC	RMSE $\downarrow$	Abs Rel $\downarrow$	$\delta_{1}$ $\uparrow$
ERP	–	–	0.4737	0.1036	88.36
Fibonacci	✓	–	0.4617	0.0968	90.62
Fibonacci	–	✓	0.4577	0.1017	90.01
ERP Grid	✓	✓	0.4558	0.0999	90.29
Fibonacci	✓	✓	0.4455	0.0949	91.16

Table 2: Ablation of node support, graph reasoning, and SCC on Matterport3D. ERP Grid uses uniformly sampled ERP nodes with matched node and edge counts.

Relations	RMSE $\downarrow$	Abs Rel $\downarrow$	$\delta_{1}$ $\uparrow$
Local	0.4577	0.0994	90.42
Local + Dilated	0.4615	0.1006	90.34
Local + Antipodal	0.4501	0.0957	90.96
Local + Dilated + Antipodal	0.4455	0.0949	91.16

Table 3: Ablation of spherical relations on Matterport3D. Fibonacci support and SCC are retained.

Support	Region	RMSE $\downarrow$	Abs Rel $\downarrow$	$\delta_{1}$ $\uparrow$
ERP Grid	Equatorial ( $30^{\circ}\mathrm{S}$ – $30^{\circ}\mathrm{N}$ )	0.5736	0.1151	87.60
Fibonacci	Equatorial ( $30^{\circ}\mathrm{S}$ – $30^{\circ}\mathrm{N}$ )	0.5585	0.1073	88.99
ERP Grid	Mid-latitude ( $30^{\circ}\mathrm{S}$ – $60^{\circ}\mathrm{S}$ , $30^{\circ}\mathrm{N}$ – $60^{\circ}\mathrm{N}$ )	0.3219	0.0894	92.19
Fibonacci	Mid-latitude ( $30^{\circ}\mathrm{S}$ – $60^{\circ}\mathrm{S}$ , $30^{\circ}\mathrm{N}$ – $60^{\circ}\mathrm{N}$ )	0.3151	0.0861	92.70
ERP Grid	High-latitude ( $60^{\circ}\mathrm{S}$ – $90^{\circ}\mathrm{S}$ , $60^{\circ}\mathrm{N}$ – $90^{\circ}\mathrm{N}$ )	0.2676	0.0822	93.29
Fibonacci	High-latitude ( $60^{\circ}\mathrm{S}$ – $90^{\circ}\mathrm{S}$ , $60^{\circ}\mathrm{N}$ – $90^{\circ}\mathrm{N}$ )	0.2605	0.0817	93.48

Table 4: Latitude-wise comparison of ERP-grid and Fibonacci node support on Matterport3D. Both variants retain the same graph reasoning and SCC configuration.

Model	FLOPs $\downarrow$	Train Mem. (GB) $\downarrow$	Inference latency $\downarrow$
UniFuse	96G	2.96	21ms
PGFuse	126G	3.63	41ms
HUSH	237G	5.11	47ms
Ours	104G	3.37	38ms

Table 5: Computational comparison at an input resolution of

512\times 1024

with a batch size of

1

, where FLOPs are computed using fvcore.nn.FlopCountAnalysis .

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Scale	$C$	$N$	$k$	Graph Blocks	$k_{b}$
$1/4$	64	2584	16	2	4
$1/8$	128	1597	12	2	4
$1/16$	256	610	8	2	2
$1/32$	512	233	8	2	2

Appendix

Table 6: Configurations of the spherical modeling modules at different feature scales.

Explore similar work

Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations
Apr 25, 2026Soulayma Gazzeh, Giuseppe Mazzola, Liliana Lo Presti +1Depth Estimation Spherical Latent Space
Department of Engineering, University of Palermo, Palermo, Italy
SO3UFormer: Learning Intrinsic Spherical Features for Rotation-Robust Panoramic Dense Prediction
Feb 26, 2026Qinfeng Zhu, Yunxi Jiang, Lei FanDense Prediction
University of Liverpool, United Kingdom. · CNRS, France. · Xi’an Jiaotong-Liverpool University, China.
SCCM: Spherically Consistent Coarse Matching for ERP Dense Feature Correspondence
Sep 29, 2026Gyeonggwan Lee, Eunsoo Im, Seunghwan Hong +1Scale-Ambiguous Epipolar Geometry Spatial Attention
Kakao Mobility Corp., Seongnam, Republic of Korea · Korea University, Seoul, Republic of Korea

Apr 25, 2026, cs.CVJ/K move · Enter open · S save

Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations

Soulayma Gazzeh, Giuseppe Mazzola, Liliana Lo Presti, Marco La Cascia

Department of Engineering, University of Palermo, Palermo, Italy

Depth Estimation Spherical Latent Space

Open paper AskPDF arXiv

Reliable depth estimation from spherical images is crucial for 360° vision in robotic navigation and immersive scene understanding. However, the onboard spherical camera can experience unintentional pose variations in real-world robotic platforms that, along with the geometric distortions inherent in equirectangular projections, significantly impact the effectiveness of depth estimation. To study this issue, a novel public benchmark, called Sphere-Depth, is introduced to systematically evaluate the robustness of monocular depth estimation models from equirectangular images in a reproducible way. Camera pose perturbations are simulated and used to assess the performance of a popular perspective-based model, Depth Anything, and of spherical-aware models such as Depth Anywhere, ACDNet, Bifuse++, and SliceNet. Furthermore, to ensure meaningful evaluation across models, a depth calibration-based error protocol is proposed to convert predicted relative depth values into metric depth values using supervised learned scaling factors for each model. Experiments show that even models explicitly designed to process spherical images exhibit substantial performance degradation when variations in the camera pose are observed with respect to the canonical pose. The full benchmark, evaluation protocol, and dataset splits are made publicly available at: https://github.com/sgazzeh/Sphere_depth

Abstract

Explore similar work

Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations

SO3UFormer: Learning Intrinsic Spherical Features for Rotation-Robust Panoramic Dense Prediction

SCCM: Spherically Consistent Coarse Matching for ERP Dense Feature Correspondence