HiP3R: Hierarchical Pose Graphs for Long-Horizon 3D Reconstruction

Papers with Code Trending Papers
3d-reconstruction pose-graph-optimization camera-trajectory long-horizon video-understanding slam

Summary

HiP3R 提出一种层级式位姿图方法，通过先优化块级 Sim(3) 图建立全局布局，再固定该布局估计每帧共享的 SE(3) 相机位姿，将局部视频重建合并为全局一致的相机轨迹和 3D 几何，无需重新训练。

Feed-forward models accurately estimate camera motion and 3D geometry from short video windows. However, computational and memory costs prevent direct inference on sequences containing thousands of frames. Long videos must therefore be processed as overlapping chunks. This division produces independent local reconstructions with inconsistent coordinate frames and duplicate estimates of shared cameras. Achieving global consistency requires propagating constraints from adjacent overlaps and distant revisits across the sequence. This requirement naturally motivates pose graph optimization. However, the choice of graph level creates a trade-off. Chunk-level graphs efficiently resolve scale drift and close large loops. Yet, their rigid transformations cannot correct internal errors or conflicting frame predictions. Conversely, frame-level optimization offers local flexibility, but scaling it globally is computationally expensive and can break metric scale. HiP3R resolves this trade-off using a hierarchical pose graph. HiP3R first optimizes a robust chunk-level Sim(3) graph to establish the global layout and scale. It then holds this layout fixed and estimates one shared SE(3) camera per frame, turning overlapping window predictions into a single trajectory. Experiments across multiple benchmarks demonstrate the effectiveness of HiP3R in improving camera trajectories and 3D reconstruction. Our method applies to both chunk-based and streaming backends without additional training or fine-tuning. The codes are available at https://kkakkkka.github.io/HiP3R.
Original Article
View Cached Full Text

Cached at: 10/28/26, 01:37 AM

Paper page - HiP3R: Hierarchical Pose Graphs for Long-Horizon 3D Reconstruction

Source: https://huggingface.co/papers/2610.31635

Abstract

HiP3R uses a hierarchical pose graph to merge local video reconstructions into globally consistent camera trajectories and 3D geometry without retraining.

Feed-forward modelsaccurately estimatecamera motionand3D geometryfrom short video windows. However, computational and memory costs prevent direct inference on sequences containing thousands of frames. Long videos must therefore be processed as overlapping chunks. This division produces independent local reconstructions with inconsistent coordinate frames and duplicate estimates of shared cameras. Achieving global consistency requires propagating constraints from adjacent overlaps and distant revisits across the sequence. This requirement naturally motivatespose graph optimization. However, the choice of graph level creates a trade-off.Chunk-level graphs efficiently resolve scale drift and close large loops. Yet, their rigid transformations cannot correct internal errors or conflicting frame predictions. Conversely,frame-level optimizationoffers local flexibility, but scaling it globally is computationally expensive and can break metric scale.HiP3Rresolves this trade-off using ahierarchical pose graph.HiP3Rfirst optimizes a robust chunk-levelSim(3)graph to establish the global layout and scale. It then holds this layout fixed and estimates one sharedSE(3)camera per frame, turning overlapping window predictions into a single trajectory. Experiments across multiple benchmarks demonstrate the effectiveness ofHiP3Rin improving camera trajectories and 3D reconstruction. Our method applies to both chunk-based and streaming backends without additional training or fine-tuning. The codes are available at https://kkakkkka.github.io/HiP3R.

View arXiv pageView PDFProject pageGitHub387Add to collection

Get this paper in your agent:

hf papers read 2610\.31635

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.31635 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.31635 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.31635 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

HRoPE: Homography Relative Positional Encoding

Hugging Face Daily Papers

HRoPE introduces a surface-aware positional encoding that models each patch as a slanted 3D plane and injects plane-induced homographies as relative transformations into attention, improving cross-view alignment and novel view synthesis without ground-truth geometry. Experiments on RealEstate10K, CO3Dv2, and DL3DV show it outperforms existing geometric positional encodings.