# Bidirectional Feature-aligned Motion Transformation for Efficient Dynamic Point Cloud Compression

arXiv: https://arxiv.org/abs/2509.14591 · HTML: https://mlsanity.com/paper/2509.14591/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 63 references that are in the map.
- First posted on arXiv 18 Sep 2025.

## Builds on (most cited first, 28 of 28)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [Joint Autoregressive and Hierarchical Priors for Learned Image Compression](https://mlsanity.com/paper/1809.02736/) (2018)
- [PointConv: Deep Convolutional Networks on 3D Point Clouds](https://mlsanity.com/paper/1811.07246/) (2018)
- [DVC: An End-to-end Deep Video Compression Framework](https://mlsanity.com/paper/1812.00101/) (2018)
- [Exploring Self-attention for Image Recognition](https://mlsanity.com/paper/2004.13621/) (2020)
- [FlowNet3D: Learning Scene Flow in 3D Point Clouds](https://mlsanity.com/paper/1806.01411/) (2018)
- [Deep Contextual Video Compression](https://mlsanity.com/paper/2109.15047/) (2021)
- [FVC: A New Framework towards Deep Video Compression in Feature Space](https://mlsanity.com/paper/2105.09600/) (2021)
- [Neural Video Compression with Feature Modulation](https://mlsanity.com/paper/2402.17414/) (2024)
- [Towards Practical Real-Time Neural Video Compression](https://mlsanity.com/paper/2502.20762/) (2025)
- [Multiscale Point Cloud Geometry Compression](https://mlsanity.com/paper/2011.03799/) (2020)
- [PV-RAFT: Point-Voxel Correlation Fields for Scene Flow Estimation of Point Clouds](https://mlsanity.com/paper/2012.00987/) (2020)
- [Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression](https://mlsanity.com/paper/2111.10633/) (2021)
- [OctAttention: Octree-Based Large-Scale Contexts Model for Point Cloud Compression](https://mlsanity.com/paper/2202.06028/) (2022)
- [Coarse-to-fine Deep Video Coding with Hyperprior-guided Mode Prediction](https://mlsanity.com/paper/2206.07460/) (2022)
- [Joint Global and Local Hierarchical Priors for Learned Image Compression](https://mlsanity.com/paper/2112.04487/) (2021)
- [Lossless Point Cloud Geometry and Attribute Compression Using a Learned Conditional Probability Model](https://mlsanity.com/paper/2303.06519/) (2023)
- [Inter-Frame Compression for Dynamic Point Cloud Geometry Coding](https://mlsanity.com/paper/2207.12554/) (2022)
- [D-DPCC: Deep Dynamic Point Cloud Compression via 3D Motion Prediction](https://mlsanity.com/paper/2205.01135/) (2022)
- [TSC-PCAC: Voxel Transformer and Sparse Convolution Based Point Cloud Attribute Compression for 3D Broadcasting](https://mlsanity.com/paper/2407.04284/) (2024)
- [Variable-Rate Learned Image Compression with Multi-Objective Optimization and Quantization-Reconstruction Offsets](https://mlsanity.com/paper/2402.18930/) (2024)
- [Learning Dynamic Point Cloud Compression via Hierarchical Inter-frame Block Matching](https://mlsanity.com/paper/2305.05356/) (2023)
- [ROI-Guided Point Cloud Geometry Compression Towards Human and Machine Vision](https://mlsanity.com/paper/2504.14240/) (2025)
- [AdaDPCC: Adaptive Rate Control and Rate-Distortion-Complexity Optimization for Dynamic Point Cloud Compression](https://mlsanity.com/paper/2508.20741/) (2025)

## Abstract (arXiv)

Efficient dynamic point cloud compression (DPCC) critically depends on accurate motion estimation and compensation. However, the inherently irregular structure and substantial local variations of point clouds make this task highly challenging. Existing approaches typically rely on explicit motion estimation, whose encoded motion vectors often fail to capture complex dynamics and inadequately exploit temporal correlations. To address these limitations, we propose a Bidirectional Feature-aligned Motion Transformation (Bi-FMT) framework that implicitly models motion in the feature space. Bi-FMT aligns features across both past and future frames to produce temporally consistent latent representations, which serve as predictive context in a conditional coding pipeline, forming a unified ``Motion + Conditional'' representation. Built upon this bidirectional feature alignment, we introduce a Cross-Transformer Refinement module (CTR) at the decoder side to adaptively refine locally aligned features. By modeling cross-frame dependencies with vector attention, CRT enhances local consistency and restores fine-grained spatial details that are often lost during motion alignment. Moreover, we design a Random Access (RA) reference strategy that treats the bidirectionally aligned features as conditional context, enabling frame-level parallel compression and eliminating the sequential encoding. Extensive experiments demonstrate that Bi-FMT surpasses D-DPCC and AdaDPCC in both compression efficiency and runtime, achieving BD-Rate reductions of 20% (D1) and 9.4% (D1), respectively.