mlsanityData to 7 Oct 2026

Real-Time Anchor-Free Single-Stage 3D Detection with IoU-Awareness

arXiv 2107.14342 ↗First posted 29 Jul 2021cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

In this report, we introduce our winning solution to the Real-time 3D Detection and also the "Most Efficient Model" in the Waymo Open Dataset Challenges at CVPR 2021. Extended from our last year's award-winning model AFDet, we have made a handful of modifications to the base model, to improve the accuracy and at the same time to greatly reduce the latency. The modified model, named as AFDetV2, is featured with a lite 3D Feature Extractor, an improved RPN with extended receptive field and an added sub-head that produces an IoU-aware confidence score. These model enhancements, together with enriched data augmentation, stochastic weights averaging, and a GPU-based implementation of voxelization, lead to a winning accuracy of 73.12 mAPH/L2 for our AFDetV2 with a latency of 60.06 ms, and an accuracy of 72.57 mAPH/L2 for our AFDetV2-base, entitled as the "Most Efficient Model" by the challenge sponsor, with a winning latency of 55.86 ms.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 22 of its 29 references that are in the map. First posted on arXiv 29 Jul 2021.

Builds on

22 of 29 references.

References · 22

  1. 2017Mask R-CNNcited by 7,259
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  3. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  4. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  5. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  6. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,194
  7. 2020Center-based 3D Object Detection and Trackingcited by 1,143
  8. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  9. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  10. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  11. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 532
  12. 2019From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Networkcited by 366
10 more
  1. 2019Class-balanced Grouping and Sampling for Point Cloud 3D Object Detectioncited by 350
  2. 2020VarifocalNet: An IoU-aware Dense Object Detectorcited by 230
  3. 2017Submanifold Sparse Convolutional Networkscited by 223
  4. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 174
  5. 2020CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloudcited by 107
  6. 2020AFDet: Anchor Free One Stage 3D Object Detectioncited by 63
  7. 2019IoU-aware Single-stage Object Detector for Accurate Localizationcited by 36
  8. 2020SWA Object Detectioncited by 22
  9. 20201st Place Solution for Waymo Open Dataset Challenge -- 3D Detection and Domain Adaptationcited by 15
  10. 20201st Place Solutions for Waymo Open Dataset Challenges -- 2D and 3D Trackingcited by 5

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3.

4 papers.

  1. 2021AFDetV2: Rethinking the Necessity of the Second Stage for Object Detection from Point Cloudscited by 88
  2. 2022Sparse Fuse Dense: Towards High Quality 3D Detection with Depth Completioncited by 82
  3. 2022CG-SSD: Corner Guided Single Stage 3D Object Detection from LiDAR Point Cloudcited by 5
  4. 2022MT-Net Submission to the Waymo 3D Detection Leaderboardcited by 5