From da2aac1db2d33d661115bd86bd25f9256dbb1283 Mon Sep 17 00:00:00 2001 From: Kaipeng Deng Date: Mon, 1 Jul 2019 11:53:29 +0800 Subject: [PATCH] update yolov3/ssd bbox AP (#2626) * update yolov3/ssd bbox AP * refine MODEL ZOO * fix AP * add size 416 320 * fix link * fix format * fix align --- docs/MODEL_ZOO.md | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/docs/MODEL_ZOO.md b/docs/MODEL_ZOO.md index 3b71432b6..b51ee3c9d 100644 --- a/docs/MODEL_ZOO.md +++ b/docs/MODEL_ZOO.md @@ -54,13 +54,20 @@ The backbone models pretrained on ImageNet are available. All backbone models ar ### Yolo v3 -| Backbone | Size | Lr schd | Box AP | Download | -| :-------- | :--: | :-----: | :----: | :-------: | -| DarkNet53 | 608 | 120e | 25.7 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_darknet.tar) | -| MobileNet-V1 | 608 | 120e | 25.7 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1.tar) | -| ResNet34 | 608 | 120e | 25.7 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1.tar) | - -- Notes: Data Augmentation（TODO：Kaipeng） +| Backbone | Size | Img/gpu | Lr schd | Box AP | Download | +| :----------- | :--: | :-----: | :-----: | :----: | :-------: | +| DarkNet53 | 608 | 8 | 120e | 38.9 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_darknet.tar) | +| DarkNet53 | 416 | 8 | 120e | 37.5 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_darknet.tar) | +| DarkNet53 | 320 | 8 | 120e | 34.8 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_darknet.tar) | +| MobileNet-V1 | 608 | 8 | 120e | 29.3 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1.tar) | +| MobileNet-V1 | 416 | 8 | 120e | 29.3 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1.tar) | +| MobileNet-V1 | 320 | 8 | 120e | 27.1 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1.tar) | +| ResNet34 | 608 | 8 | 120e | 36.2 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_r34.tar) | +| ResNet34 | 416 | 8 | 120e | 34.3 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_r34.tar) | +| ResNet34 | 320 | 8 | 120e | 31.4 | [model](https://paddlemodels.bj.bcebos.com/object_detection/yolov3_r34.tar) | + +**NOTE**: Yolo v3 trained in 8 GPU with total batch size as 64. Yolo v3 training data augmentations: mixup image, +random distort image, random crop image, random expand image, random interpolate, random flip image. ### RetinaNet @@ -69,10 +76,11 @@ The backbone models pretrained on ImageNet are available. All backbone models ar | ResNet50-FPN | 300 | 120e | 36.0 | [model](https://paddlemodels.bj.bcebos.com/object_detection/retinanet_r50_fpn_1x.tar) | | ResNet101-FPN | 300 | 120e | 37.3 | [model](https://paddlemodels.bj.bcebos.com/object_detection/retinanet_r101_fpn_1x.tar) | -- Notes: （TODO：Kaipeng） - ### SSD on PascalVOC -| Backbone | Size | Lr schd | Box AP | Download | -| :----------- | :--: | :-----: | :----: | :-------: | -| MobileNet v1 | 300 | 120e | 25.7 | [model](https://paddlemodels.bj.bcebos.com/object_detection/ssd_mobilenet_v1_voc.tar) | +| Backbone | Size | Img/gpu | Lr schd | Box AP | Download | +| :----------- | :--: | :-----: | :-----: | :----: | :-------: | +| MobileNet v1 | 300 | 32 | 120e | 73.2 | [model](https://paddlemodels.bj.bcebos.com/object_detection/ssd_mobilenet_v1_voc.tar) | + +**NOTE**: SSD trained in 2 GPU with totoal batch size as 64. SSD training data augmentations: random distort image, +random crop image, random expand image, random flip image. -- GitLab