# Objects are Different: Flexible Monocular 3D Object Detection
> [Objects are Different: Flexible Monocular 3D Object Detection](https://arxiv.org/abs/2104.02323)
<!-- [ALGORITHM] -->
## Abstract
The precise localization of 3D objects from a single image without depth information is a highly challenging problem. Most existing methods adopt the same approach for all objects regardless of their diverse distributions, leading to limited performance for truncated objects. In this paper, we propose a flexible framework for monocular 3D object detection which explicitly decouples the truncated objects and adaptively combines multiple approaches for object depth estimation. Specifically, we decouple the edge of the feature map for predicting long-tail truncated objects so that the optimization of normal objects is not influenced. Furthermore, we formulate the object depth estimation as an uncertainty-guided ensemble of directly regressed object depth and solved depths from different groups of keypoints. Experiments demonstrate that our method outperforms the state-of-the-art method by relatively 27% for the moderate level and 30% for the hard level in the test set of KITTI benchmark while maintaining real-time efficiency.
Note: mAP represents Car moderate 3D strict AP11 / AP40 results. Because of the limited data for pedestrians and cyclists, the detection performance for these two classes is usually unstable. Therefore, we only list car detection results here. In addition, the AP11 result may fluctuate in a larger range (~1 AP), so AP40 is a more recommended metric for reference due to its much better stability.
## Citation
```latex
@InProceedings{MonoFlex,
author = {Zhang, Yunpeng and Lu, Jiwen and Zhou, Jie},
title = {Objects Are Different: Flexible Monocular 3D Object Detection},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
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 [this https URL](https://github.com/Tai-Wang/Depth-from-Motion).
Regrettably, we are unable to provide the pre-trained model weights due to [Waymo Dataset License Agreement](https://waymo.com/open/terms/), so we only provide the training logs as shown above.
## Citation
```latex
@article{wang2022mvfcos3d++,
title={{MV-FCOS3D++: Multi-View} Camera-Only 4D Object Detection with Pretrained Monocular Backbones},
author={Wang, Tai and Lian, Qing and Zhu, Chenming and Zhu, Xinge and Zhang, Wenwei},
# MVX-Net: Multimodal VoxelNet for 3D Object Detection
> [MVX-Net: Multimodal VoxelNet for 3D Object Detection](https://arxiv.org/abs/1904.01649)
<!-- [ALGORITHM] -->
## Abstract
Many recent works on 3D object detection have focused on designing neural network architectures that can consume point cloud data. While these approaches demonstrate encouraging performance, they are typically based on a single modality and are unable to leverage information from other modalities, such as a camera. Although a few approaches fuse data from different modalities, these methods either use a complicated pipeline to process the modalities sequentially, or perform late-fusion and are unable to learn interaction between different modalities at early stages. In this work, we present PointFusion and VoxelFusion: two simple yet effective early-fusion approaches to combine the RGB and point cloud modalities, by leveraging the recently introduced VoxelNet architecture. Evaluation on the KITTI dataset demonstrates significant improvements in performance over approaches which only use point cloud data. Furthermore, the proposed method provides results competitive with the state-of-the-art multimodal algorithms, achieving top-2 ranking in five of the six bird's eye view and 3D detection categories on the KITTI benchmark, by using a simple single stage network.
We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance. The method, called Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN, running at 5 fps. Moreover, Mask R-CNN is easy to generalize to other tasks, e.g., allowing us to estimate human poses in the same framework. We show top results in all three tracks of the COCO suite of challenges, including instance segmentation, bounding-box object detection, and person keypoint detection. Without bells and whistles, Mask R-CNN outperforms all existing, single-model entries on every task, including the COCO 2016 challenge winners. We hope our simple and effective approach will serve as a solid baseline and help ease future research in instance-level recognition.
We support and provide some baseline results on [nuImages dataset](https://www.nuscenes.org/nuimages).
We follow the class mapping in nuScenes dataset, which maps the original categories into 10 foreground categories.
The convert script can be found [here](https://github.com/open-mmlab/mmdetection3d/blob/master/tools/dataset_converters/nuimage_converter.py).
The baseline results include instance segmentation models, e.g., Mask R-CNN, Cascade Mask R-CNN, and HTC.
We will support panoptic segmentation models in the future.
The dataset converted by the script of v0.6.0 only supports instance segmentation. Since v0.7.0, we also support to produce semantic segmentation mask of each image; thus, we can train HTC or semantic segmentation models using the dataset. To convert the nuImages dataset into COCO format, please use the command below:
-`--data-root`: the root of the dataset, defaults to `./data/nuimages`.
-`--version`: the version of the dataset, defaults to `v1.0-mini`. To get the full dataset, please use `--version v1.0-train v1.0-val v1.0-mini`
-`--out-dir`: the output directory of annotations and semantic masks, defaults to `./data/nuimages/annotations/`.
-`--nproc`: number of workers for data preparation, defaults to `4`. Larger number could reduce the preparation time as images are processed in parallel.
-`--extra-tag`: extra tag of the annotations, defaults to `nuimages`. This can be used to separate different annotations processed in different time for study.
## Results and models
### Instance Segmentation
We report Mask R-CNN and Cascade Mask R-CNN results on nuimages.
| Method | Backbone | Pretraining | Lr schd | Mem (GB) | Box AP | Mask AP | Download |
1.`IN` means only using ImageNet pre-trained backbone. `IN+COCO-Nx` and `IN+COCO-Ne` means the backbone is first pre-trained on ImageNet, and then the detector is pre-trained on COCO train2017 dataset by `Nx` and `N` epochs schedules, respectively.
2. All the training hyper-parameters follow the standard schedules on COCO dataset except that the images are resized from
1280 x 720 to 1920 x 1080 (relative ratio 0.8 to 1.2) since the images are in size 1600 x 900.
3. The class order in the detectors released in v0.6.0 is different from the order in the configs because the bug in the conversion script. This bug has been fixed since v0.7.0 and the models trained by the correct class order are also released. If you used nuImages since v0.6.0, please re-convert the data through the conversion script using the above-mentioned command.
