MMDetection & object detection方面的调研

MMDetection(2D)

MMDetection is an open source object detection toolbox based on PyTorch. It consists of:

• Training recipes for object detection and instance segmentation.
• 360+ pre-trained models to use for fine-tuning 微调 (or training afresh).
• Dataset support for popular vision datasets such as COCO, Cityscapes, LVIS and PASCAL VOC.

https://paperswithcode.com/lib/mmdetection

	MODEL	TRANING ON	BOX AP	INFERENCETIME	PAPER	YEAR
	Dynamic R-CNN	COCO	38.9			2020
	DETR	COCO	40.1			2020
	GRoIE	COCO	42.2			2020
	RegNet	COCO	43.1			2020
	PAA	COCO	45.1			2020
	Sparse R-CNN	COCO	46.2			2020
	ResNeSt	COCO	47.7			2020
	DetectoRS	COCO	49.1			2020
	VFNet	COCO	50.4			2020
	Generalized Focal Loss	COCO	48.1	0.0935		2020

https://github.com/open-mmlab/mmdetection3d

[

OpenMMLab website HOT OpenMMLab platform TRY IT OUT

MMDetection3D

MMDetection3D is an open source object detection toolbox based on PyTorch, towards the next-generation platform for general 3D detection. It is a part of the OpenMMLab project developed by MMLab.（支持分布式训练和推理；it supports distributed training and inference.）

Major features

• Support multi-modality/single-modality detectors out of box

  It directly supports multi-modality(多模态)/single-modality(单模态) detectors including MVXNet, VoteNet, PointPillars, etc.

• Support indoor/outdoor 3D detection out of box

  It directly supports popular indoor and outdoor 3D detection datasets, including ScanNet, SUNRGB-D, Waymo, nuScenes, Lyft, and KITTI. For nuScenes dataset, we also support nuImages dataset.

• Natural integration with 2D detection

  All the about 300+ models, methods of 40+ papers, and modules supported in MMDetection can be trained or used in this codebase.

• High efficiency

  It trains faster than other codebases. The main results are as below. Details can be found in benchmark.md. We compare the number of samples trained per second (the higher, the better). The models that are not supported by other codebases are marked by ✗.

  Methods	MMDetection3D	OpenPCDet	votenet	Det3D
  VoteNet	358	✗	77	✗
  PointPillars-car	141	✗	✗	140
  PointPillars-3class	107	44	✗	✗
  SECOND	40	30	✗	✗
  Part-A2	17	14	✗	✗

Like MMDetection and MMCV, MMDetection3D can also be used as a library to support different projects on top of it.

Benchmarks

Here we benchmark the training and testing speed of models in MMDetection3D, with some other open source 3D detection codebases.

Settings

• Hardwares: 8 NVIDIA Tesla V100 (32G) GPUs, Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
  Software: Python 3.7, CUDA 10.1, cuDNN 7.6.5, PyTorch 1.3, numba 0.48.0.
• Model: Since all the other codebases implements different models, we compare the corresponding models including SECOND, PointPillars, Part-A2, and VoteNet with them separately.
• Metrics: We use the average throughput in iterations of the entire training run and skip the first 50 iterations of each epoch to skip GPU warmup time.

Main Results

We compare the training speed (samples/s) with other codebases if they implement the similar models. The results are as below, the greater the numbers in the table, the faster of the training process. The models that are not supported by other codebases are marked by ×.

Methods	MMDetection3D	OpenPCDet	votenet	Det3D
VoteNet	358	×	77	×
PointPillars-car	141	×	×	140
PointPillars-3class	107	44	×	×
SECOND	40	30	×	×
Part-A2	17	14	×	×

In summary, MMdetection is a toolbox for object detection, MMCV is a foundation library(基础库) for computer vision research that supports MMDetection, and MMDetection3D is an extension of MMDetection for 3D object detection and tracking.

总之，MMDetection是一个物体检测的工具箱，MMCV是一个支持MMDetection的计算机视觉研究的基础库，而MMDetection3D是MMDetection的扩展，用于3D物体检测和跟踪。

Verify the installation

# single-gpu testing
python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--cfg-options test_evaluator.pklfile_prefix=${RESULT_FILE}]  [--show] [--show-dir ${SHOW_DIR}]

# e.g.
python demo/pcd_demo.py demo/data/kitti/000008.bin pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py hv_pointpillars_secfpn_6x8_160e_kitti-3d-car_20220331_134606-d42d15ed.pth --show

python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth --show --show-dir ./data/scannet/show_results

• VoteNet

• PointPillars

• SECOND

• Part-A2

雷达参数

目前大车使用的镭神智能的雷达参数

测绘激光雷达三维测绘激光雷达 MS-C16

大场景快速建模	近程高精建图

镭神智能128 线混合固态激光雷达 CH128 效果

非机动车	轿车