# Ultra-High Resolution Segmentation via Boundary-Enhanced Patch-Merging Transformer

arXiv: https://arxiv.org/abs/2412.10181 · HTML: https://mlsanity.com/paper/2412.10181/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 73 references that are in the map.
- First posted on arXiv 13 Dec 2024.

## Builds on (most cited first, 12 of 12)

- [DFormer: Rethinking RGBD Representation Learning for Semantic Segmentation](https://mlsanity.com/paper/2309.09668/) (2023)
- [CamoFormer: Masked Separable Attention for Camouflaged Object Detection](https://mlsanity.com/paper/2212.06570/) (2022)
- [MetaGPT: Merging Large Language Models Using Model Exclusive Task Arithmetic](https://mlsanity.com/paper/2406.11385/) (2024)
- [DrivingRecon: Large 4D Gaussian Reconstruction Model For Autonomous Driving](https://mlsanity.com/paper/2412.09043/) (2024)
- [SGLP: A Similarity Guided Fast Layer Partition Pruning for Compressing Large Deep Models](https://mlsanity.com/paper/2410.14720/) (2024)
- [MSP-MVS: Multi-Granularity Segmentation Prior Guided Multi-View Stereo](https://mlsanity.com/paper/2407.19323/) (2024)
- [Guided Patch-Grouping Wavelet Transformer with Spatial Congruence for Ultra-High Resolution Segmentation](https://mlsanity.com/paper/2307.00711/) (2023)
- [COMAE: COMprehensive Attribute Exploration for Zero-shot Hashing](https://mlsanity.com/paper/2402.16424/) (2024)
- [Geolocation Representation from Large Language Models are Generic Enhancers for Spatio-Temporal Learning](https://mlsanity.com/paper/2408.12116/) (2024)
- [Triplet Contrastive Representation Learning for Unsupervised Vehicle Re-identification](https://mlsanity.com/paper/2301.09498/) (2023)
- [Scaling Multi-Camera 3D Object Detection through Weak-to-Strong Eliciting](https://mlsanity.com/paper/2404.06700/) (2024)
- [Exact: Exploring Space-Time Perceptive Clues for Weakly Supervised Satellite Image Time Series Semantic Segmentation](https://mlsanity.com/paper/2412.03968/) (2024)

## Built on by (most cited first, 10 of 10)

- [Pinwheel-shaped Convolution and Scale-based Dynamic Loss for Infrared Small Target Detection](https://mlsanity.com/paper/2412.16986/) (2024)
- [DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation](https://mlsanity.com/paper/2504.04701/) (2025)
- [From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective](https://mlsanity.com/paper/2503.13165/) (2025)
- [Conditional Latent Coding with Learnable Synthesized Reference for Deep Image Compression](https://mlsanity.com/paper/2502.09971/) (2025)
- [From Zero to Detail: A Progressive Spectral Decoupling Paradigm for UHD Image Restoration with New Benchmark](https://mlsanity.com/paper/2604.15654/) (2026)
- [Scale-adaptive UAV Geo-localization via Height-aware Partition Learning](https://mlsanity.com/paper/2412.11535/) (2024)
- [STeInFormer: Spatial-Temporal Interaction Transformer Architecture for Remote Sensing Change Detection](https://mlsanity.com/paper/2412.17247/) (2024)
- [Granular Computing-driven SAM: From Coarse-to-Fine Guidance for Prompt-Free Segmentation](https://mlsanity.com/paper/2511.19062/) (2025)
- [Adapting Vision Transformers to Ultra-High Resolution Semantic Segmentation with Relay Tokens](https://mlsanity.com/paper/2601.05927/) (2026)
- [SFR-Net: Learning Scale-Frustum Representations for Ultra-Wide Area Remote Sensing Image Segmentation](https://mlsanity.com/paper/2605.25737/) (2026)

## Abstract (arXiv)

Segmentation of ultra-high resolution (UHR) images is a critical task with numerous applications, yet it poses significant challenges due to high spatial resolution and rich fine details. Recent approaches adopt a dual-branch architecture, where a global branch learns long-range contextual information and a local branch captures fine details. However, they struggle to handle the conflict between global and local information while adding significant extra computational cost. Inspired by the human visual system's ability to rapidly orient attention to important areas with fine details and filter out irrelevant information, we propose a novel UHR segmentation method called Boundary-enhanced Patch-merging Transformer (BPT). BPT consists of two key components: (1) Patch-Merging Transformer (PMT) for dynamically allocating tokens to informative regions to acquire global and local representations, and (2) Boundary-Enhanced Module (BEM) that leverages boundary information to enrich fine details. Extensive experiments on multiple UHR image segmentation benchmarks demonstrate that our BPT outperforms previous state-of-the-art methods without introducing extra computational overhead. Codes will be released to facilitate research.