PanoSeg3R: Feed-Forward 3D Semantic Segmentation for Panoramic Images with an Automatic Data Curation Pipeline
cs.CV
Submitted: 2026-09-19
Updated: 2026-09-19
Project page: https://harryyoon777.github.io/PanoSeg3R/Introduction
Terminology
Sources
- Joint 2D-3D-Semantic Data for Indoor Scene Understanding
- Qwen3-VL Technical Report
- SAM 3: Segment Anything with Concepts
- PanoSAMic: Panoramic Image Segmentation from SAM Feature Encoding and Dual View Fusion
- Spherical CNNs on Unstructured Grids
- Wid3R: Wide Field-of-View 3D Reconstruction via Camera Model Conditioning
- Unified Semantic Transformer for 3D Scene Understanding
- Language-driven Semantic Segmentation
- IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstruction
- SGAT4PASS: Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentation
- Decoupled Weight Decay Regularization
- SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
- $\pi^3$: Permutation-Equivariant Visual Geometry Learning
- MTPano: Multi-Task Panoramic Scene Understanding via Label-Free Integration of Dense Prediction Priors
- Panoramic Scene Understanding: A Survey from Distortion-Aware Engineering to Sphere-Native Modeling
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models