移动paddle_detection

2024-09-24 17:02:56 +08:00
parent 90a6d5ec75
commit 3438cf6e0e
2025 changed files with 11 additions and 11 deletions
--- a/services/paddle_services/paddle_detection/configs/fcos/README.md
+++ b/services/paddle_services/paddle_detection/configs/fcos/README.md
@@ -0,0 +1,37 @@
+# FCOS (Fully Convolutional One-Stage Object Detection)
+
+## Model Zoo on COCO
+
+| 骨架网络             | 网络类型       | 每张GPU图片个数 | 学习率策略 |推理时间(fps) | Box AP |                           下载                          | 配置文件 |
+| :------------------- | :------------- | :-----: | :-----: | :------------: | :-----: | :-----------------------------------------------------: | :-----: |
+| ResNet50-FPN    | FCOS           |    2    |   1x      |     ----     |  39.6  | [download](https://paddledet.bj.bcebos.com/models/fcos_r50_fpn_1x_coco.pdparams) | [config](./fcos_r50_fpn_1x_coco.yml) |
+| ResNet50-FPN    | FCOS + iou      |    2    |   1x      |     ----     |  40.0  | [download](https://paddledet.bj.bcebos.com/models/fcos_r50_fpn_iou_1x_coco.pdparams) | [config](./fcos_r50_fpn_iou_1x_coco.yml) |
+| ResNet50-FPN    | FCOS + DCN       |    2    |   1x      |     ----     |  44.3  | [download](https://paddledet.bj.bcebos.com/models/fcos_dcn_r50_fpn_1x_coco.pdparams) | [config](./fcos_dcn_r50_fpn_1x_coco.yml) |
+| ResNet50-FPN    | FCOS + multiscale_train    |    2    |   2x      |     ----     |  41.8  | [download](https://paddledet.bj.bcebos.com/models/fcos_r50_fpn_multiscale_2x_coco.pdparams) | [config](./fcos_r50_fpn_multiscale_2x_coco.yml) |
+| ResNet50-FPN    | FCOS + multiscale_train + iou    |    2    |   2x      |     ----     |  42.6  | [download](https://paddledet.bj.bcebos.com/models/fcos_r50_fpn_iou_multiscale_2x_coco.pdparams) | [config](./fcos_r50_fpn_iou_multiscale_2x_coco.yml) |
+
+**注意:**
+  - `+ iou` 表示与原版 FCOS 相比，不使用 `centerness` 而是使用 `iou` 来参与计算loss。
+  - 基于 FCOS 的半监督检测方法 `DenseTeaher` 可以参照[DenseTeaher](../semi_det/denseteacher)去使用，结合无标签数据可以进一步提升检测性能。
+  - PaddleDetection中默认使用`R50-vb`预训练，如果使用`R50-vd`结合[SSLD](../../../docs/feature_models/SSLD_PRETRAINED_MODEL.md)的预训练模型，可进一步显著提升检测精度，同时backbone部分配置也需要做出相应更改，如：
+  ```python
+    pretrain_weights: https://paddledet.bj.bcebos.com/models/pretrained/ResNet50_vd_ssld_v2_pretrained.pdparams
+    ResNet:
+      depth: 50
+      variant: d
+      norm_type: bn
+      freeze_at: 0
+      return_idx: [1, 2, 3]
+      num_stages: 4
+      lr_mult_list: [0.05, 0.05, 0.1, 0.15]
+  ```
+
+## Citations
+```
+@inproceedings{tian2019fcos,
+  title   =  {{FCOS}: Fully Convolutional One-Stage Object Detection},
+  author  =  {Tian, Zhi and Shen, Chunhua and Chen, Hao and He, Tong},
+  booktitle =  {Proc. Int. Conf. Computer Vision (ICCV)},
+  year    =  {2019}
+}
+```
--- a/services/paddle_services/paddle_detection/configs/fcos/_base_/fcos_r50_fpn.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/_base_/fcos_r50_fpn.yml
@@ -0,0 +1,48 @@
+architecture: FCOS
+pretrain_weights: https://paddledet.bj.bcebos.com/models/pretrained/ResNet50_cos_pretrained.pdparams
+
+FCOS:
+  backbone: ResNet
+  neck: FPN
+  fcos_head: FCOSHead
+
+ResNet:
+  depth: 50
+  variant: 'b'
+  norm_type: bn
+  freeze_at: 0 # res2
+  return_idx: [1, 2, 3]
+  num_stages: 4
+
+FPN:
+  out_channel: 256
+  spatial_scales: [0.125, 0.0625, 0.03125]
+  extra_stage: 2
+  has_extra_convs: True
+  use_c5: False
+
+FCOSHead:
+  fcos_feat:
+    name: FCOSFeat
+    feat_in: 256
+    feat_out: 256
+    num_convs: 4
+    norm_type: "gn"
+    use_dcn: False
+  fpn_stride: [8, 16, 32, 64, 128]
+  prior_prob: 0.01
+  norm_reg_targets: True
+  centerness_on_reg: True
+  num_shift: 0.5
+  fcos_loss:
+    name: FCOSLoss
+    loss_alpha: 0.25
+    loss_gamma: 2.0
+    iou_loss_type: "giou"
+    reg_weights: 1.0
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.025
+    nms_threshold: 0.6
--- a/services/paddle_services/paddle_detection/configs/fcos/_base_/fcos_reader.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/_base_/fcos_reader.yml
@@ -0,0 +1,42 @@
+worker_num: 2
+TrainReader:
+  sample_transforms:
+  - Decode: {}
+  - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+  - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+  - RandomFlip: {}
+  batch_transforms:
+  - Permute: {}
+  - PadBatch: {pad_to_stride: 128}
+  - Gt2FCOSTarget:
+      object_sizes_boundary: [64, 128, 256, 512]
+      center_sampling_radius: 1.5
+      downsample_ratios: [8, 16, 32, 64, 128]
+      norm_reg_targets: True
+  batch_size: 2
+  shuffle: True
+  drop_last: True
+  use_shared_memory: True
+
+
+EvalReader:
+  sample_transforms:
+  - Decode: {}
+  - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+  - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+  - Permute: {}
+  batch_transforms:
+  - PadBatch: {pad_to_stride: 128}
+  batch_size: 1
+
+
+TestReader:
+  sample_transforms:
+  - Decode: {}
+  - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+  - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+  - Permute: {}
+  batch_transforms:
+  - PadBatch: {pad_to_stride: 128}
+  batch_size: 1
+  fuse_normalize: True
--- a/services/paddle_services/paddle_detection/configs/fcos/_base_/optimizer_1x.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/_base_/optimizer_1x.yml
@@ -0,0 +1,19 @@
+epoch: 12
+
+LearningRate:
+  base_lr: 0.01
+  schedulers:
+  - !PiecewiseDecay
+    gamma: 0.1
+    milestones: [8, 11]
+  - !LinearWarmup
+    start_factor: 0.3333333333333333
+    steps: 500
+
+OptimizerBuilder:
+  optimizer:
+    momentum: 0.9
+    type: Momentum
+  regularizer:
+    factor: 0.0001
+    type: L2
--- a/services/paddle_services/paddle_detection/configs/fcos/fcos_dcn_r50_fpn_1x_coco.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/fcos_dcn_r50_fpn_1x_coco.yml
@@ -0,0 +1,16 @@
+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '_base_/fcos_r50_fpn.yml',
+  '_base_/optimizer_1x.yml',
+  '_base_/fcos_reader.yml',
+]
+
+weights: output/fcos_dcn_r50_fpn_1x_coco/model_final
+
+ResNet:
+  dcn_v2_stages: [1, 2, 3]
+
+FCOSHead:
+  fcos_feat:
+    use_dcn: True
--- a/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_1x_coco.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_1x_coco.yml
@@ -0,0 +1,9 @@
+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '_base_/fcos_r50_fpn.yml',
+  '_base_/optimizer_1x.yml',
+  '_base_/fcos_reader.yml',
+]
+
+weights: output/fcos_r50_fpn_1x_coco/model_final
--- a/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_iou_1x_coco.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_iou_1x_coco.yml
@@ -0,0 +1,79 @@
+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '_base_/fcos_r50_fpn.yml',
+  '_base_/optimizer_1x.yml',
+  '_base_/fcos_reader.yml',
+]
+
+weights: output/fcos_r50_fpn_iou_1x_coco/model_final
+
+
+TrainReader:
+  sample_transforms:
+    - Decode: {}
+    - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - RandomFlip: {}
+  batch_transforms:
+    - Permute: {}
+    - PadBatch: {pad_to_stride: 32}
+    - Gt2FCOSTarget:
+        object_sizes_boundary: [64, 128, 256, 512]
+        center_sampling_radius: 1.5
+        downsample_ratios: [8, 16, 32, 64, 128]
+        norm_reg_targets: True
+  batch_size: 2
+  shuffle: True
+  drop_last: True
+  use_shared_memory: True
+
+
+EvalReader:
+  sample_transforms:
+    - Decode: {}
+    - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - Permute: {}
+  batch_transforms:
+    - PadBatch: {pad_to_stride: 32}
+  batch_size: 1
+
+
+TestReader:
+  sample_transforms:
+    - Decode: {}
+    - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - Permute: {}
+  batch_transforms:
+    - PadBatch: {pad_to_stride: 32}
+  batch_size: 1
+  fuse_normalize: True
+
+
+FCOSHead:
+  fcos_feat:
+    name: FCOSFeat
+    feat_in: 256
+    feat_out: 256
+    num_convs: 4
+    norm_type: "gn"
+    use_dcn: False
+  fpn_stride: [8, 16, 32, 64, 128]
+  prior_prob: 0.01
+  norm_reg_targets: True
+  centerness_on_reg: True
+  fcos_loss:
+    name: FCOSLoss
+    loss_alpha: 0.25
+    loss_gamma: 2.0
+    iou_loss_type: "giou"
+    reg_weights: 1.0
+    quality: "iou" # default 'centerness'
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.025
+    nms_threshold: 0.6
--- a/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_iou_multiscale_2x_coco.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_iou_multiscale_2x_coco.yml
@@ -0,0 +1,91 @@
+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '_base_/fcos_r50_fpn.yml',
+  '_base_/optimizer_1x.yml',
+  '_base_/fcos_reader.yml',
+]
+
+weights: output/fcos_r50_fpn_iou_multiscale_2x_coco_010/model_final
+
+TrainReader:
+  sample_transforms:
+    - Decode: {}
+    - RandomResize: {target_size: [[640, 1333], [672, 1333], [704, 1333], [736, 1333], [768, 1333], [800, 1333]], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - RandomFlip: {}
+  batch_transforms:
+    - Permute: {}
+    - PadBatch: {pad_to_stride: 32}
+    - Gt2FCOSTarget:
+        object_sizes_boundary: [64, 128, 256, 512]
+        center_sampling_radius: 1.5
+        downsample_ratios: [8, 16, 32, 64, 128]
+        norm_reg_targets: True
+  batch_size: 2
+  shuffle: True
+  drop_last: True
+  use_shared_memory: True
+
+
+EvalReader:
+  sample_transforms:
+    - Decode: {}
+    - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - Permute: {}
+  batch_transforms:
+    - PadBatch: {pad_to_stride: 32}
+  batch_size: 1
+
+
+TestReader:
+  sample_transforms:
+    - Decode: {}
+    - Resize: {target_size: [800, 1333], keep_ratio: True, interp: 1}
+    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+    - Permute: {}
+  batch_transforms:
+    - PadBatch: {pad_to_stride: 32}
+  batch_size: 1
+  fuse_normalize: True
+
+
+epoch: 24
+
+LearningRate:
+  base_lr: 0.01
+  schedulers:
+  - !PiecewiseDecay
+    gamma: 0.1
+    milestones: [16, 22]
+  - !LinearWarmup
+    start_factor: 0.001
+    steps: 1000
+
+
+FCOSHead:
+  fcos_feat:
+    name: FCOSFeat
+    feat_in: 256
+    feat_out: 256
+    num_convs: 4
+    norm_type: "gn"
+    use_dcn: False
+  fpn_stride: [8, 16, 32, 64, 128]
+  prior_prob: 0.01
+  norm_reg_targets: True
+  centerness_on_reg: True
+  fcos_loss:
+    name: FCOSLoss
+    loss_alpha: 0.25
+    loss_gamma: 2.0
+    iou_loss_type: "giou"
+    reg_weights: 1.0
+    quality: "iou" # default 'centerness'
+  nms:
+    name: MultiClassNMS
+    nms_top_k: 1000
+    keep_top_k: 100
+    score_threshold: 0.025
+    nms_threshold: 0.6
--- a/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_multiscale_2x_coco.yml
+++ b/services/paddle_services/paddle_detection/configs/fcos/fcos_r50_fpn_multiscale_2x_coco.yml
@@ -0,0 +1,40 @@
+_BASE_: [
+  '../datasets/coco_detection.yml',
+  '../runtime.yml',
+  '_base_/fcos_r50_fpn.yml',
+  '_base_/optimizer_1x.yml',
+  '_base_/fcos_reader.yml',
+]
+
+weights: output/fcos_r50_fpn_multiscale_2x_coco/model_final
+
+TrainReader:
+  sample_transforms:
+  - Decode: {}
+  - RandomResize: {target_size: [[640, 1333], [672, 1333], [704, 1333], [736, 1333], [768, 1333], [800, 1333]], keep_ratio: True, interp: 1}
+  - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
+  - RandomFlip: {}
+  batch_transforms:
+  - Permute: {}
+  - PadBatch: {pad_to_stride: 128}
+  - Gt2FCOSTarget:
+      object_sizes_boundary: [64, 128, 256, 512]
+      center_sampling_radius: 1.5
+      downsample_ratios: [8, 16, 32, 64, 128]
+      norm_reg_targets: True
+  batch_size: 2
+  shuffle: True
+  drop_last: True
+  use_shared_memory: True
+
+epoch: 24
+
+LearningRate:
+  base_lr: 0.01
+  schedulers:
+  - !PiecewiseDecay
+    gamma: 0.1
+    milestones: [16, 22]
+  - !LinearWarmup
+    start_factor: 0.3333333333333333
+    steps: 500