NAS-FCOS

Model Name:*

Description with Markdown (optional):

# NAS-FCOS: Fast Neural Architecture Search for Object Detection

## Introduction

[ALGORITHM]

```latex
@article{wang2019fcos,
  title={Nas-fcos: Fast neural architecture search for object detection},
  author={Wang, Ning and Gao, Yang and Chen, Hao and Wang, Peng and Tian, Zhi and Shen, Chunhua},
  journal={arXiv preprint arXiv:1906.04423},
  year={2019}
}
```

## Results and Models

| Head      | Backbone  | Style   | GN-head | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download |
|:---------:|:---------:|:-------:|:-------:|:-------:|:--------:|:--------------:|:------:|:------:|:--------:|
| NAS-FCOSHead | R-50   | caffe   | Y       | 1x      |          |                | 39.4   | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/nas_fcos/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco.py) | [model](http://download.openmmlab.com/mmdetection/v2.0/nas_fcos/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco_20200520-1bdba3ce.pth) &#124; [log](http://download.openmmlab.com/mmdetection/v2.0/nas_fcos/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco/nas_fcos_nashead_r50_caffe_fpn_gn-head_4x4_1x_coco_20200520.log.json) |
| FCOSHead  | R-50      | caffe   | Y       | 1x      |          |                | 38.5   | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/nas_fcos/nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco.py) | [model](http://download.openmmlab.com/mmdetection/v2.0/nas_fcos/nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco/nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco_20200521-7fdcbce0.pth) &#124; [log](http://download.openmmlab.com/mmdetection/v2.0/nas_fcos/nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco/nas_fcos_fcoshead_r50_caffe_fpn_gn-head_4x4_1x_coco_20200521.log.json) |

**Notes:**

- To be consistent with the author's implementation, we use 4 GPUs with 4 images/GPU.

Paper:*

Code URL (optional):

lr sched	1x
Backbone Layers	50

lr sched	1x
Backbone Layers	50

FCOS

BENCHMARK	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
COCO minival	NAS-FCOSHead (R-50, NAS-FCOSHead, 1x, caffe)	box AP	39.4	# 80
COCO minival	FCOSHead (R-50, FCOSHead, 1x, caffe)	box AP	38.5	# 89

Architecture	Non Maximum Suppression, NAS-FCOS, ResNet, FPN
lr sched	1x
Backbone Layers	50
SHOW MORE
SHOW LESS

Architecture	Non Maximum Suppression, ResNet, FPN, FCOS
lr sched	1x
Backbone Layers	50
SHOW MORE
SHOW LESS

Head	Backbone	Style	GN-head	Lr schd	Mem (GB)	Inf time (fps)	box AP	Config	Download
NAS-FCOSHead	R-50	caffe	Y	1x			39.4	config	model \| log
FCOSHead	R-50	caffe	Y	1x			38.5	config	model \| log

open-mmlab / mmdetection

NAS-FCOS: Fast Neural Architecture Search for Object Detection

Introduction

Results and Models

Results

Object Detection