DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers

• URL: http://arxiv.org/abs/2606.31585v1
• Date: Tue, 30 Jun 2026 12:38:58 GMT
• Title: DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers
• Abstract summary: Camera-based positional encoding is crucial for providing spatial cues in 3D computer vision.<n>In this paper, we propose a novel camera-based positional encoding that explicitly decouples rotation and translation.<n>Extensive evaluations on NVS tasks demonstrate that DPPE enables stable long-term training even in scaled-up training setup.
• Score: 4.777114198703681
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The remarkable scalability of Transformers has expanded their application to 3D computer vision, where camera-aware positional encoding is crucial for providing spatial cues in multi-view geometry. Recent advancements have established the practice of using camera parameters -- such as extrinsics or projection matrices -- as relative positional encoding into the query, key, and value vectors of the attention mechanism. However, when scaling up the training recipe of novel view synthesis (NVS) models with the camera-based positional encoding, we observe a significant issue: model performance stagnates in the late stages of training. In this paper, we investigate the cause of the performance bottleneck when scaling up and demonstrate that storing rotation and translation given by the positional encoding in the same dimensions of the value vector causes indeterminacy in their independent identification, hindering training scalability. To address this, we propose Decoupled Pose Positional Encoding (DPPE), a novel camera-based positional encoding that explicitly decouples rotation and translation. Extensive evaluations on NVS tasks demonstrate that DPPE enables stable long-term training even in scaled-up training setup. Furthermore, it exhibits superior generalization performance in extrapolation settings, such as handling an increased number of viewpoints and zoom-in scenarios.

Related papers

• Unified Camera Positional Encoding for Controlled Video Generation [48.5789182990001]
  Transformers have emerged as a universal backbone across 3D perception, video generation, and world models for autonomous driving and embodied AI.<n>We introduce Relative Ray, a geometry-consistent representation that unifies complete camera information, including 6-DoF poses, intrinsics, and lens distortions.<n>To facilitate systematic training and evaluation, we construct a large video dataset covering a wide range of camera motions and lens types.
  arXiv  Detail & Related papers  (2025-12-08T07:34:01Z)
• Mitigating Coordinate Prediction Bias from Positional Encoding Failures [31.63900089866925]
  High-resolution inputs exacerbate this difficulty by producing long token sequences that weaken positional encodings.<n>We analyze how MLLMs behave when visual positional encodings are deliberately perturbed through shuffling.<n>We propose Vision-PE Shuffle Guidance (VPSG), a training-free test-time method that leverages the directional nature of these biases for correction.
  arXiv  Detail & Related papers  (2025-10-25T00:58:47Z)
• FreqPDE: Rethinking Positional Depth Embedding for Multi-View 3D Object Detection Transformers [91.59069344768858]
  We introduce Frequency-aware Positional Depth Embedding (FreqPDE) to equip 2D image features with spatial information for 3D detection transformer decoder.<n>FreqPDE combines the 2D image features and 3D position embeddings to generate 3D depth-aware features for query decoding.
  arXiv  Detail & Related papers  (2025-10-17T07:36:54Z)
• Cameras as Relative Positional Encoding [37.675563572777136]
  Multi-view transformers must use camera geometry to ground visual tokens in 3D space.<n>We show how relative camera conditioning improves performance in feedforward novel view synthesis.<n>We then verify that these benefits persist for different tasks, stereo depth estimation and discriminative cognition, as well as larger model sizes.
  arXiv  Detail & Related papers  (2025-07-14T17:22:45Z)
• SeqPE: Transformer with Sequential Position Encoding [76.22159277300891]
  SeqPE represents each $n$-dimensional position index as a symbolic sequence and employs a lightweight sequential position encoder to learn their embeddings.<n> Experiments across language modeling, long-context question answering, and 2D image classification demonstrate that SeqPE not only surpasses strong baselines in perplexity, exact match (EM) and accuracy--but also enables seamless generalization to multi-dimensional inputs without requiring manual architectural redesign.
  arXiv  Detail & Related papers  (2025-06-16T09:16:40Z)
• Toward Relative Positional Encoding in Spiking Transformers [76.72869420863749]
  Spiking neural networks (SNNs) are bio-inspired networks that mimic how neurons in the brain communicate through discrete spikes.<n>We introduce several strategies to approximate relative positional encoding in spiking Transformers while preserving the binary nature of spikes.
  arXiv  Detail & Related papers  (2025-01-28T06:42:37Z)
• Learnable Fourier Features for Multi-DimensionalSpatial Positional Encoding [96.9752763607738]
  We propose a novel positional encoding method based on learnable Fourier features. Our experiments show that our learnable feature representation for multi-dimensional positional encoding outperforms existing methods.
  arXiv  Detail & Related papers  (2021-06-05T04:40:18Z)
• TransCamP: Graph Transformer for 6-DoF Camera Pose Estimation [77.09542018140823]
  We propose a neural network approach with a graph transformer backbone, namely TransCamP, to address the camera relocalization problem. TransCamP effectively fuses the image features, camera pose information and inter-frame relative camera motions into encoded graph attributes.
  arXiv  Detail & Related papers  (2021-05-28T19:08:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.