add vitdet_faster_rcnn (#155)

* add vitdet_faster_Rcn and refactor vitdet_config

configs/detection/fcos/coco_detection.py

@@ -88,3 +88,14 @@ val_dataset = dict(
 
 data = dict(
     imgs_per_gpu=2, workers_per_gpu=2, train=train_dataset, val=val_dataset)
+
+# evaluation
+eval_config = dict(interval=1, gpu_collect=False)
+eval_pipelines = [
+    dict(
+        mode='test',
+        evaluators=[
+            dict(type='CocoDetectionEvaluator', classes=CLASSES),
+        ],
+    )
+]

configs/detection/fcos/fcos_center-normbbox-centeronreg-giou_r50_caffe_fpn_gn-head_1x_coco.py

@@ -1,21 +1,5 @@
 _base_ = ['./fcos.py', './coco_detection.py', 'configs/base.py']
 
-CLASSES = [
-    'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
-    'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
-    'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
-    'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
-    'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
-    'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
-    'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
-    'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
-    'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
-    'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
-    'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
-    'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
-    'hair drier', 'toothbrush'
-]
-
 log_config = dict(
     interval=50,
     hooks=[
@@ -42,15 +26,4 @@ lr_config = dict(
 
 total_epochs = 12
 
-# evaluation
-eval_config = dict(interval=1, gpu_collect=False)
-eval_pipelines = [
-    dict(
-        mode='test',
-        evaluators=[
-            dict(type='CocoDetectionEvaluator', classes=CLASSES),
-        ],
-    )
-]
-
 find_unused_parameters = False

configs/detection/vitdet/_base_/datasets/coco_detection.py

@@ -0,0 +1,115 @@
+CLASSES = [
+    'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
+    'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
+    'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
+    'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
+    'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
+    'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
+    'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
+    'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
+    'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
+    'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
+    'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
+    'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
+    'hair drier', 'toothbrush'
+]
+
+# dataset settings
+data_root = 'data/coco/'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+
+image_size = (1024, 1024)
+train_pipeline = [
+    # large scale jittering
+    dict(
+        type='MMResize',
+        img_scale=image_size,
+        ratio_range=(0.1, 2.0),
+        multiscale_mode='range',
+        keep_ratio=True),
+    dict(
+        type='MMRandomCrop',
+        crop_type='absolute_range',
+        crop_size=image_size,
+        recompute_bbox=False,
+        allow_negative_crop=True),
+    dict(type='MMFilterAnnotations', min_gt_bbox_wh=(1e-2, 1e-2)),
+    dict(type='MMRandomFlip', flip_ratio=0.5),
+    dict(type='MMNormalize', **img_norm_cfg),
+    dict(type='MMPad', size=image_size),
+    dict(type='DefaultFormatBundle'),
+    dict(
+        type='Collect',
+        keys=['img', 'gt_bboxes', 'gt_labels'],
+        meta_keys=('filename', 'ori_filename', 'ori_shape', 'ori_img_shape',
+                   'img_shape', 'pad_shape', 'scale_factor', 'flip',
+                   'flip_direction', 'img_norm_cfg'))
+]
+test_pipeline = [
+    dict(
+        type='MMMultiScaleFlipAug',
+        img_scale=image_size,
+        flip=False,
+        transforms=[
+            dict(type='MMResize', keep_ratio=True),
+            dict(type='MMRandomFlip'),
+            dict(type='MMNormalize', **img_norm_cfg),
+            dict(type='MMPad', size_divisor=1024),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(
+                type='Collect',
+                keys=['img'],
+                meta_keys=('filename', 'ori_filename', 'ori_shape',
+                           'ori_img_shape', 'img_shape', 'pad_shape',
+                           'scale_factor', 'flip', 'flip_direction',
+                           'img_norm_cfg'))
+        ])
+]
+
+train_dataset = dict(
+    type='DetDataset',
+    data_source=dict(
+        type='DetSourceCoco',
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='LoadAnnotations', with_bbox=True)
+        ],
+        classes=CLASSES,
+        test_mode=False,
+        filter_empty_gt=True,
+        iscrowd=False),
+    pipeline=train_pipeline)
+
+val_dataset = dict(
+    type='DetDataset',
+    imgs_per_gpu=1,
+    data_source=dict(
+        type='DetSourceCoco',
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=[
+            dict(type='LoadImageFromFile'),
+            dict(type='LoadAnnotations', with_bbox=True)
+        ],
+        classes=CLASSES,
+        test_mode=True,
+        filter_empty_gt=False,
+        iscrowd=True),
+    pipeline=test_pipeline)
+
+data = dict(
+    imgs_per_gpu=1, workers_per_gpu=2, train=train_dataset, val=val_dataset)
+
+# evaluation
+eval_config = dict(interval=1, gpu_collect=False)
+eval_pipelines = [
+    dict(
+        mode='test',
+        evaluators=[
+            dict(type='CocoDetectionEvaluator', classes=CLASSES),
+        ],
+    )
+]

configs/detection/vitdet/_base_/datasets/coco_instance.py

@@ -102,3 +102,15 @@ val_dataset = dict(
 
 data = dict(
     imgs_per_gpu=1, workers_per_gpu=2, train=train_dataset, val=val_dataset)
+
+# evaluation
+eval_config = dict(interval=1, gpu_collect=False)
+eval_pipelines = [
+    dict(
+        mode='test',
+        evaluators=[
+            dict(type='CocoDetectionEvaluator', classes=CLASSES),
+            dict(type='CocoMaskEvaluator', classes=CLASSES)
+        ],
+    )
+]

configs/detection/vitdet/_base_/models/vitdet_faster_rcnn.py

@@ -0,0 +1,126 @@
+# model settings
+
+norm_cfg = dict(type='GN', num_groups=1, requires_grad=True)
+
+pretrained = 'https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/selfsup/mae/vit-b-1600/warpper_mae_vit-base-p16-1600e.pth'
+model = dict(
+    type='FasterRCNN',
+    pretrained=pretrained,
+    backbone=dict(
+        type='ViTDet',
+        img_size=1024,
+        embed_dim=768,
+        depth=12,
+        num_heads=12,
+        mlp_ratio=4,
+        qkv_bias=True,
+        qk_scale=None,
+        drop_rate=0.,
+        attn_drop_rate=0.,
+        drop_path_rate=0.1,
+        use_abs_pos_emb=True,
+        aggregation='attn',
+    ),
+    neck=dict(
+        type='SFP',
+        in_channels=[768, 768, 768, 768],
+        out_channels=256,
+        norm_cfg=norm_cfg,
+        num_outs=5),
+    rpn_head=dict(
+        type='RPNHead',
+        in_channels=256,
+        feat_channels=256,
+        num_convs=2,
+        norm_cfg=norm_cfg,
+        anchor_generator=dict(
+            type='AnchorGenerator',
+            scales=[8],
+            ratios=[0.5, 1.0, 2.0],
+            strides=[4, 8, 16, 32, 64]),
+        bbox_coder=dict(
+            type='DeltaXYWHBBoxCoder',
+            target_means=[.0, .0, .0, .0],
+            target_stds=[1.0, 1.0, 1.0, 1.0]),
+        loss_cls=dict(
+            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
+        loss_bbox=dict(type='L1Loss', loss_weight=1.0)),
+    roi_head=dict(
+        type='StandardRoIHead',
+        bbox_roi_extractor=dict(
+            type='SingleRoIExtractor',
+            roi_layer=dict(type='RoIAlign', output_size=7, sampling_ratio=0),
+            out_channels=256,
+            featmap_strides=[4, 8, 16, 32]),
+        bbox_head=dict(
+            type='Shared4Conv1FCBBoxHead',
+            conv_out_channels=256,
+            norm_cfg=norm_cfg,
+            in_channels=256,
+            fc_out_channels=1024,
+            roi_feat_size=7,
+            num_classes=80,
+            bbox_coder=dict(
+                type='DeltaXYWHBBoxCoder',
+                target_means=[0., 0., 0., 0.],
+                target_stds=[0.1, 0.1, 0.2, 0.2]),
+            reg_class_agnostic=False,
+            loss_cls=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
+            loss_bbox=dict(type='L1Loss', loss_weight=1.0))),
+    # model training and testing settings
+    train_cfg=dict(
+        rpn=dict(
+            assigner=dict(
+                type='MaxIoUAssigner',
+                pos_iou_thr=0.7,
+                neg_iou_thr=0.3,
+                min_pos_iou=0.3,
+                match_low_quality=True,
+                ignore_iof_thr=-1),
+            sampler=dict(
+                type='RandomSampler',
+                num=256,
+                pos_fraction=0.5,
+                neg_pos_ub=-1,
+                add_gt_as_proposals=False),
+            allowed_border=-1,
+            pos_weight=-1,
+            debug=False),
+        rpn_proposal=dict(
+            nms_pre=2000,
+            max_per_img=1000,
+            nms=dict(type='nms', iou_threshold=0.7),
+            min_bbox_size=0),
+        rcnn=dict(
+            assigner=dict(
+                type='MaxIoUAssigner',
+                pos_iou_thr=0.5,
+                neg_iou_thr=0.5,
+                min_pos_iou=0.5,
+                match_low_quality=True,
+                ignore_iof_thr=-1),
+            sampler=dict(
+                type='RandomSampler',
+                num=512,
+                pos_fraction=0.25,
+                neg_pos_ub=-1,
+                add_gt_as_proposals=True),
+            pos_weight=-1,
+            debug=False)),
+    test_cfg=dict(
+        rpn=dict(
+            nms_pre=1000,
+            max_per_img=1000,
+            nms=dict(type='nms', iou_threshold=0.7),
+            min_bbox_size=0),
+        rcnn=dict(
+            score_thr=0.05,
+            nms=dict(type='nms', iou_threshold=0.5),
+            max_per_img=100)))
+
+mmlab_modules = [
+    dict(type='mmdet', name='FasterRCNN', module='model'),
+    dict(type='mmdet', name='RPNHead', module='head'),
+    dict(type='mmdet', name='StandardRoIHead', module='head'),
+]

configs/detection/vitdet/vitdet_100e.py

@@ -3,22 +3,6 @@ _base_ = [
     'configs/base.py'
 ]
 
-CLASSES = [
-    'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
-    'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
-    'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
-    'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
-    'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
-    'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
-    'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
-    'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
-    'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
-    'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
-    'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
-    'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
-    'hair drier', 'toothbrush'
-]
-
 log_config = dict(
     interval=50,
     hooks=[
@@ -50,16 +34,4 @@ lr_config = dict(
     step=[88, 96])
 total_epochs = 100
 
-# evaluation
-eval_config = dict(interval=1, gpu_collect=False)
-eval_pipelines = [
-    dict(
-        mode='test',
-        evaluators=[
-            dict(type='CocoDetectionEvaluator', classes=CLASSES),
-            dict(type='CocoMaskEvaluator', classes=CLASSES)
-        ],
-    )
-]
-
 find_unused_parameters = False

configs/detection/vitdet/vitdet_basicblock_100e.py

@@ -1,67 +1,3 @@
-_base_ = [
+_base_ = './vitdet_100e.py'
-    './_base_/models/vitdet.py', './_base_/datasets/coco_instance.py',
-    'configs/base.py'
-]
-
-CLASSES = [
-    'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
-    'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
-    'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
-    'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
-    'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
-    'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
-    'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
-    'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
-    'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
-    'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
-    'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
-    'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
-    'hair drier', 'toothbrush'
-]
 
 model = dict(backbone=dict(aggregation='basicblock'))
-
-log_config = dict(
-    interval=50,
-    hooks=[
-        dict(type='TextLoggerHook'),
-        # dict(type='TensorboardLoggerHook')
-    ])
-
-checkpoint_config = dict(interval=10)
-# optimizer
-paramwise_options = {
-    'norm': dict(weight_decay=0.),
-    'bias': dict(weight_decay=0.),
-    'pos_embed': dict(weight_decay=0.),
-    'cls_token': dict(weight_decay=0.)
-}
-optimizer = dict(
-    type='AdamW',
-    lr=1e-4,
-    betas=(0.9, 0.999),
-    weight_decay=0.1,
-    paramwise_options=paramwise_options)
-optimizer_config = dict(grad_clip=None, loss_scale=512.)
-# learning policy
-lr_config = dict(
-    policy='step',
-    warmup='linear',
-    warmup_iters=250,
-    warmup_ratio=0.067,
-    step=[88, 96])
-total_epochs = 100
-
-# evaluation
-eval_config = dict(interval=1, gpu_collect=False)
-eval_pipelines = [
-    dict(
-        mode='test',
-        evaluators=[
-            dict(type='CocoDetectionEvaluator', classes=CLASSES),
-            dict(type='CocoMaskEvaluator', classes=CLASSES)
-        ],
-    )
-]
-
-find_unused_parameters = False

configs/detection/vitdet/vitdet_bottleneck_100e.py

@@ -1,67 +1,3 @@
-_base_ = [
+_base_ = './vitdet_100e.py'
-    './_base_/models/vitdet.py', './_base_/datasets/coco_instance.py',
-    'configs/base.py'
-]
-
-CLASSES = [
-    'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
-    'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
-    'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
-    'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
-    'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
-    'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
-    'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
-    'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
-    'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
-    'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
-    'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
-    'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
-    'hair drier', 'toothbrush'
-]
 
 model = dict(backbone=dict(aggregation='bottleneck'))
-
-log_config = dict(
-    interval=50,
-    hooks=[
-        dict(type='TextLoggerHook'),
-        # dict(type='TensorboardLoggerHook')
-    ])
-
-checkpoint_config = dict(interval=10)
-# optimizer
-paramwise_options = {
-    'norm': dict(weight_decay=0.),
-    'bias': dict(weight_decay=0.),
-    'pos_embed': dict(weight_decay=0.),
-    'cls_token': dict(weight_decay=0.)
-}
-optimizer = dict(
-    type='AdamW',
-    lr=1e-4,
-    betas=(0.9, 0.999),
-    weight_decay=0.1,
-    paramwise_options=paramwise_options)
-optimizer_config = dict(grad_clip=None, loss_scale=512.)
-# learning policy
-lr_config = dict(
-    policy='step',
-    warmup='linear',
-    warmup_iters=250,
-    warmup_ratio=0.067,
-    step=[88, 96])
-total_epochs = 100
-
-# evaluation
-eval_config = dict(interval=1, gpu_collect=False)
-eval_pipelines = [
-    dict(
-        mode='test',
-        evaluators=[
-            dict(type='CocoDetectionEvaluator', classes=CLASSES),
-            dict(type='CocoMaskEvaluator', classes=CLASSES)
-        ],
-    )
-]
-
-find_unused_parameters = False

configs/detection/vitdet/vitdet_faster_rcnn_100e.py

@@ -0,0 +1,37 @@
+_base_ = [
+    './_base_/models/vitdet_faster_rcnn.py',
+    './_base_/datasets/coco_detection.py', 'configs/base.py'
+]
+
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+checkpoint_config = dict(interval=10)
+# optimizer
+paramwise_options = {
+    'norm': dict(weight_decay=0.),
+    'bias': dict(weight_decay=0.),
+    'pos_embed': dict(weight_decay=0.),
+    'cls_token': dict(weight_decay=0.)
+}
+optimizer = dict(
+    type='AdamW',
+    lr=1e-4,
+    betas=(0.9, 0.999),
+    weight_decay=0.1,
+    paramwise_options=paramwise_options)
+optimizer_config = dict(grad_clip=None, loss_scale=512.)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=250,
+    warmup_ratio=0.067,
+    step=[88, 96])
+total_epochs = 100
+
+find_unused_parameters = False