# MV-FCOS3D++: Multi-View Camera-Only 4D Object Detection with Pretrained Monocular Backbones

Source: [MV-FCOS3D++: Multi-View Camera-Only 4D Object Detection with Pretrained Monocular Backbones](<https://www.sota2.com/research/paper/68fa6d1e1e209131cea37e5e>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:26:36.450Z

Authors: Tai Wang, Qing Lian, Chenming Zhu, Xinge Zhu, Wenwei Zhang

Published: 2022-07-26T08:10:29.000Z

## Abstract

In this technical report, we present our solution, dubbed MV-FCOS3D++, for the Camera-Only 3D Detection track in Waymo Open Dataset Challenge 2022. For multi-view camera-only 3D detection, methods based on bird-eye-view or 3D geometric representations can leverage the stereo cues from overlapped regions between adjacent views and directly perform 3D detection without hand-crafted post-processing. However, it lacks direct semantic supervision for 2D backbones, which can be complemented by pretraining simple monocular-based detectors. Our solution is a multi-view framework for 4D detection following this paradigm. It is built upon a simple monocular detector FCOS3D++, pretrained only with object annotations of Waymo, and converts multi-view features to a 3D grid space to detect 3D objects thereon. A dual-path neck for single-frame understanding and temporal stereo matching is devised to incorporate multi-frame information. Our method finally achieves 49.75% mAPL with a single model and wins 2nd place in the WOD challenge, without any LiDAR-based depth supervision during training. The code will be released at https://github.com/Tai-Wang/Depth-from-Motion.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Multi-camera 3D Object Detection on Waymo Open Dataset (WOD) v1.3.1 (val)](<https://www.sota2.com/research/sota/multi-camera-3d-object-detection-on-waymo-open-dataset-wod-v1-3-1-val.md>): Multi-camera 3D Object Detection; Waymo Open Dataset (WOD) v1.3.1 (val); LET-APL: 58.11.

- [3D Object Detection on Waymo Open Dataset (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-val.md>): 3D Object Detection; Waymo Open Dataset (val); 3D APH Vehicle L2: not recorded.

- [3D Object Detection on Waymo (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-val.md>): 3D Object Detection; Waymo (val); Vehicle L2 AP: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2207.12716>)

[PDF](<https://arxiv.org/pdf/2207.12716.pdf>)