search

zhengye1995 / Zero-shot-Instance-Segmentation

code for CVPR paper Zero-shot Instance Segmentation
Apache License 2.0
93 stars 14 forks source link

how are the difference between bbox_head_semantic and convfc_bbox_semantic_head? #18

Closed algoteam5 closed 3 years ago

algoteam5 commented 3 years ago

I am confusing about your work.

  1. how is the difference between bbox_head_semantic and convfc_bbox_semantic_head?
  2. how is the difference between ba_anchor_head and anchor_semantic_head?
  3. how are the difference among zero_shot_faster_rcnn, zero_shot_mask_rcnn and zero_shot_two_stage ?
zhengye1995 commented 3 years ago
  1. convfc_bbox_semantic_head inherits bbox_head_semantic likes convfc_bbox_head inherits bbox_head in mmdetection
  2. ba_rpn_head inherits ba_anchor_head
  3. zero_shot_faster_rcnn and zero_shot_mask_rcnn inherit zero_shot_two_stage
algoteam5 commented 3 years ago

@zhengye1995 I know they are inherited. What I am asking is that what is the use of each one? i.e. why did you use zero_shot_mask_rcnn instead of zero_shot_faster_rcnn in the config?

zhengye1995 commented 3 years ago

zero_shot_faster_rcnn is used for ZSD task and zero_shot_mask_rcnn is used for ZSI task. The biggest difference is whether they include the mask branch, see this code.

By the way, this is just my personal implementation habit. You can also add the mask branch into zero_shot_faster_rcnn to merge the two implementations into one.

algoteam5 commented 3 years ago

Could you please give me the config file where ZSD is run only? so that I can understand how was your work integrated inside the mmdetection framework. I tried like this but got errors


# model settings
model = dict(
    type='ZeroShotFasterRCNN',
    pretrained='torchvision://resnet101',
    backbone=dict(
        type='ResNet',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        style='pytorch'),
    neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        num_outs=5),
    rpn_head=dict(
        type='BackgroundAwareRPNHead',
        in_channels=256,
        semantic_dims=300,
        feat_channels=256,
        anchor_scales=[8],
        anchor_ratios=[0.5, 1.0, 2.0],
        anchor_strides=[4, 8, 16, 32, 64],
        target_means=[.0, .0, .0, .0],
        target_stds=[1.0, 1.0, 1.0, 1.0],
        voc_path=None,
        vec_path='data/coco/word_w2v_withbg_65_15.txt',
        sync_bg=True,
        loss_cls=dict(
            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
        loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, loss_weight=1.0)),
    bbox_roi_extractor=dict(
        type='SingleRoIExtractor',
        roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
        out_channels=256,
        featmap_strides=[4, 8, 16, 32]),
    bbox_head=dict(
        type='SharedFCSemanticBBoxHead',
        num_fcs=2,
        in_channels=256,
        fc_out_channels=1024,
        roi_feat_size=7,
        num_classes=66,
        semantic_dims=300,
        seen_class=True,
        reg_with_semantic=False,
        share_semantic=False,
        with_decoder=True,
        sync_bg=True,
        voc_path='data/coco/vocabulary_w2v.txt',
        vec_path='data/coco/word_w2v_withbg_65_15.txt',
        target_means=[0., 0., 0., 0.],
        target_stds=[0.1, 0.1, 0.2, 0.2],
        reg_class_agnostic=False,
        loss_semantic=dict(
            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
        loss_bbox=dict(type='SmoothL1Loss', beta=1.0, 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,
            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=0,
        pos_weight=-1,
        debug=False),
    rpn_proposal=dict(
        nms_across_levels=False,
        nms_pre=2000,
        nms_post=2000,
        max_num=2000,
        nms_thr=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,
            ignore_iof_thr=-1),
        sampler=dict(
            type='RandomSampler',
            num=512,
            pos_fraction=0.25,
            neg_pos_ub=-1,
            add_gt_as_proposals=True),
        mask_size=28,
        pos_weight=-1,
        debug=False))

test_cfg = dict(
    rpn=dict(
        nms_across_levels=False,
        nms_pre=1000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        score_thr=0.05,
        nms=dict(type='nms', iou_thr=0.5),
        max_per_img=100,
        mask_thr_binary=0.5))

# dataset settings
dataset_type = 'CocoDatasetUnseen15'
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)

train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
    dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='Pad', size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
]

test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(1333, 800),
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize', **img_norm_cfg),
            dict(type='Pad', size_divisor=32),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img']),
        ])
]

data = dict(
    imgs_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_train2014_seen_65_15.json',
        img_prefix=data_root + 'train2014/',
        pipeline=train_pipeline),
    val=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_val2014_seen_65_15.json',
        img_prefix=data_root + 'val2014/',
        pipeline=test_pipeline),
    test=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_val2014_unseen_65_15.json',
        img_prefix=data_root + 'val2014/',
        pipeline=test_pipeline))

# optimizer
optimizer = dict(type='SGD', lr=0.0075, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))

# learning policy
lr_config = dict(
    policy='step',
    warmup='linear',
    warmup_iters=7000,
    warmup_ratio=1.0 / 3,
    step=[8, 11])
checkpoint_config = dict(interval=12)

# yapf:disable
log_config = dict(
    interval=50,
    hooks=[
        dict(type='TextLoggerHook'),
        # dict(type='TensorboardLoggerHook')
    ])

# yapf:enable
evaluation = dict(interval=1)

# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/zsi/baseline/65_15/'
load_from = None
resume_from = None
workflow = [('train', 1)]
zhengye1995 commented 3 years ago

Is it convenient for you to provide detailed information about the error?

algoteam5 commented 3 years ago

Not at all : ), youre very polite! here is the error:


Traceback (most recent call last):
  File "./tools/train.py", line 119, in <module>
    main()
  File "./tools/train.py", line 115, in main
    logger=logger)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/apis/train.py", line 58, in train_detector
    _dist_train(model, dataset, cfg, validate=validate)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/apis/train.py", line 192, in _dist_train
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs)
  File "/home/usdd1/anaconda3/envs/zsi2_cu11/lib/python3.7/site-packages/mmcv/runner/runner.py", line 358, in run
    epoch_runner(data_loaders[i], **kwargs)
  File "/home/usdd1/anaconda3/envs/zsi2_cu11/lib/python3.7/site-packages/mmcv/runner/runner.py", line 264, in train
    self.model, data_batch, train_mode=True, **kwargs)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/apis/train.py", line 38, in batch_processor
    losses = model(**data)
  File "/home/usdd1/anaconda3/envs/zsi2_cu11/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/usdd1/anaconda3/envs/zsi2_cu11/lib/python3.7/site-packages/mmcv/parallel/distributed.py", line 50, in forward
    return self.module(*inputs[0], **kwargs[0])
  File "/home/usdd1/anaconda3/envs/zsi2_cu11/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/core/fp16/decorators.py", line 49, in new_func
    return old_func(*args, **kwargs)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/models/detectors/base.py", line 117, in forward
    return self.forward_train(img, img_meta, **kwargs)
  File "/home/usdd1/yahee/instance_seg/ZSI2_cu11/mmdet/models/detectors/zero_shot_two_stage.py", line 193, in forward_train
    *rpn_loss_inputs, gt_bboxes_ignore=gt_bboxes_ignore)
TypeError: loss() got multiple values for argument 'gt_bboxes_ignore'
zhengye1995 commented 3 years ago

Try this one:

# model settings
model = dict(
    type='ZeroShotFasterRCNN',
    pretrained='torchvision://resnet101',
    backbone=dict(
        type='ResNet',
        depth=101,
        num_stages=4,
        out_indices=(0, 1, 2, 3),
        frozen_stages=1,
        style='pytorch'),
    neck=dict(
        type='FPN',
        in_channels=[256, 512, 1024, 2048],
        out_channels=256,
        num_outs=5),
    rpn_head=dict(
        type='BackgroundAwareRPNHead',
        in_channels=256,
        semantic_dims=300,
        feat_channels=256,
        anchor_scales=[8],
        anchor_ratios=[0.5, 1.0, 2.0],
        anchor_strides=[4, 8, 16, 32, 64],
        target_means=[.0, .0, .0, .0],
        target_stds=[1.0, 1.0, 1.0, 1.0],
        voc_path=None,
        vec_path='data/coco/word_w2v_withbg_65_15.txt',
        sync_bg=True,
        loss_cls=dict(
            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
        loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, loss_weight=1.0)),
    bbox_roi_extractor=dict(
        type='SingleRoIExtractor',
        roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
        out_channels=256,
        featmap_strides=[4, 8, 16, 32]),
    bbox_head=dict(
        type='SharedFCSemanticBBoxHead',
        num_fcs=2,
        in_channels=256,
        fc_out_channels=1024,
        roi_feat_size=7,
        num_classes=66,
        semantic_dims=300,
        seen_class=True,
        reg_with_semantic=False,
        share_semantic=False,
        with_decoder=True,
        sync_bg=True,
        voc_path='data/coco/vocabulary_w2v.txt',
        vec_path='data/coco/word_w2v_withbg_65_15.txt',
        target_means=[0., 0., 0., 0.],
        target_stds=[0.1, 0.1, 0.2, 0.2],
        reg_class_agnostic=False,
        loss_semantic=dict(
            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
        loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
    bbox_with_decoder=True,
    bbox_sync_bg=True)
# 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,
            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=0,
        pos_weight=-1,
        debug=False),
    rpn_proposal=dict(
        nms_across_levels=False,
        nms_pre=2000,
        nms_post=2000,
        max_num=2000,
        nms_thr=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,
            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_across_levels=False,
        nms_pre=1000,
        nms_post=1000,
        max_num=1000,
        nms_thr=0.7,
        min_bbox_size=0),
    rcnn=dict(
        score_thr=0.05,
        nms=dict(type='nms', iou_thr=0.5),
        max_per_img=100))
# dataset settings
dataset_type = 'CocoDatasetUnseen15'
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)
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations', with_bbox=True),
    dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='Pad', size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']),
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(1333, 800),
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize', **img_norm_cfg),
            dict(type='Pad', size_divisor=32),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img']),
        ])
]
data = dict(
    imgs_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_train2014_seen_65_15.json',
        img_prefix=data_root + 'train2014/',
        pipeline=train_pipeline),
    val=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_val2014_seen_65_15.json',
        img_prefix=data_root + 'val2014/',
        pipeline=test_pipeline),
    test=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_val2014_unseen_65_15.json',
        img_prefix=data_root + 'val2014/',
        pipeline=test_pipeline))
# optimizer
optimizer = dict(type='SGD', lr=0.0075, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
    policy='step',
    warmup='linear',
    warmup_iters=7000,
    warmup_ratio=1.0 / 3,
    step=[8, 11])
checkpoint_config = dict(interval=12)
# yapf:disable
log_config = dict(
    interval=50,
    hooks=[
        dict(type='TextLoggerHook'),
        # dict(type='TensorboardLoggerHook')
    ])
# yapf:enable
evaluation = dict(interval=1)
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/zsi/baseline/65_15/'
load_from = None
resume_from = None
workflow = [('train', 1)]
algoteam5 commented 3 years ago

Thank you!