# QAL: A Loss for Recall Precision Balance in 3D Reconstruction

arXiv: https://arxiv.org/abs/2511.17824 · HTML: https://mlsanity.com/paper/2511.17824/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 41 references that are in the map.
- First posted on arXiv 21 Nov 2025.

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

- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection](https://mlsanity.com/paper/1711.06396/) (2017)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [A Point Set Generation Network for 3D Object Reconstruction from a Single Image](https://mlsanity.com/paper/1612.00603/) (2016)
- [PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud](https://mlsanity.com/paper/1812.04244/) (2018)
- [Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images](https://mlsanity.com/paper/1804.01654/) (2018)
- [Deep Hough Voting for 3D Object Detection in Point Clouds](https://mlsanity.com/paper/1904.09664/) (2019)
- [3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction](https://mlsanity.com/paper/1604.00449/) (2016)
- [PCN: Point Completion Network](https://mlsanity.com/paper/1808.00671/) (2018)
- [PIXOR: Real-time 3D Object Detection from Point Clouds](https://mlsanity.com/paper/1902.06326/) (2019)
- [PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers](https://mlsanity.com/paper/2108.08839/) (2021)
- [What Do Single-view 3D Reconstruction Networks Learn?](https://mlsanity.com/paper/1905.03678/) (2019)
- [Pix2Vox: Context-aware 3D Reconstruction from Single and Multi-view Images](https://mlsanity.com/paper/1901.11153/) (2019)
- [PF-Net: Point Fractal Network for 3D Point Cloud Completion](https://mlsanity.com/paper/2003.00410/) (2020)
- [Density-aware Chamfer Distance as a Comprehensive Metric for Point Cloud Completion](https://mlsanity.com/paper/2111.12702/) (2021)
- [UniDexGrasp++: Improving Dexterous Grasping Policy Learning via Geometry-aware Curriculum and Iterative Generalist-Specialist Learning](https://mlsanity.com/paper/2304.00464/) (2023)
- [Variational Relational Point Completion Network](https://mlsanity.com/paper/2104.10154/) (2021)
- [PMP-Net: Point Cloud Completion by Learning Multi-step Point Moving Paths](https://mlsanity.com/paper/2012.03408/) (2020)
- [SeedFormer: Patch Seeds based Point Cloud Completion with Upsample Transformer](https://mlsanity.com/paper/2207.10315/) (2022)
- [PointNetGPD: Detecting Grasp Configurations from Point Sets](https://mlsanity.com/paper/1809.06267/) (2018)
- [Shape Completion Enabled Robotic Grasping](https://mlsanity.com/paper/1609.08546/) (2016)
- [Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Images](https://mlsanity.com/paper/2006.12250/) (2020)
- [LegoFormer: Transformers for Block-by-Block Multi-view 3D Reconstruction](https://mlsanity.com/paper/2106.12102/) (2021)
- [BUOL: A Bottom-Up Framework with Occupancy-aware Lifting for Panoptic 3D Scene Reconstruction From A Single Image](https://mlsanity.com/paper/2306.00965/) (2023)
- [Beyond Top-Grasps Through Scene Completion](https://mlsanity.com/paper/1909.12908/) (2019)

## Abstract (arXiv)

Volumetric learning underpins many 3D vision tasks such as completion, reconstruction, and mesh generation, yet training objectives still rely on Chamfer Distance (CD) or Earth Mover's Distance (EMD), which fail to balance recall and precision. We propose Quality-Aware Loss (QAL), a drop-in replacement for CD/EMD that combines a coverage-weighted nearest-neighbor term with an uncovered-ground-truth attraction term, explicitly decoupling recall and precision into tunable components. Across diverse pipelines, QAL achieves consistent coverage gains, improving by an average of +4.3 pts over CD and +2.8 pts over the best alternatives. Though modest in percentage, these improvements reliably recover thin structures and under-represented regions that CD/EMD overlook. Extensive ablations confirm stable performance across hyperparameters and across output resolutions, while full retraining on PCN and ShapeNet demonstrates generalization across datasets and backbones. Moreover, QAL-trained completions yield higher grasp scores under GraspNet evaluation, showing that improved coverage translates directly into more reliable robotic manipulation. QAL thus offers a principled, interpretable, and practical objective for robust 3D vision and safety-critical robotics pipelines