add detection code

README.md

@@ -19,8 +19,8 @@ The architecture of GC ViT is demonstrated in the following:
 ![teaser](./assets/gc_vit.png)
 
 ## 💥 News 💥
-
-- **[05.21.2023]** 🔥🔥 We have released ImageNet-21K fine-tuned GC ViT model weights for 224x224 and 384x384.
+- **[07.21.2023]** 🔥 We have released the object detection/instance segmentation ![code](./detection/README.md) ! 
+- **[05.21.2023]** 🔥 We have released ImageNet-21K fine-tuned GC ViT model weights for 224x224 and 384x384.
 - **[05.21.2023]** 🔥🔥 We have released new ImageNet-1K GC ViT model weights with **better performance** !
 - **[04.24.2023]** 🔥🔥🔥 GC ViT has been accepted to **ICML 2023** !
 

detection/README.md

@@ -0,0 +1,23 @@
+# GC ViT - Object Detection
+This repository is the official PyTorch implementation of Global Context Vision Transformers for object detection using MS COCO dataset.
+
+## Requirements
+The dependencies can be installed by running:
+
+```bash
+pip install -r requirements.txt
+```
+
+
+## Benchmarks
+
+The expected performance of models that use GC ViT as a backbone is listed below:
+
+| Backbone | Head | #Params(M) | FLOPs(G) | mAP | Mask mAP|
+|---|---|---|---|---|---|
+| GC ViT-T | Mask R-CNN | 48 | 291 | 47.9 | 43.2 |
+| GC ViT-T | Cascade Mask R-CNN | 85 | 770 | 51.6 | 44.6 |
+| GC ViT-S | Cascade Mask R-CNN | 108 | 866 | 52.4 | 45.4 |
+| GC ViT-B | Cascade Mask R-CNN | 146 | 1018 | 52.9 | 45.8 |
+
+

detection/configs/_base_/datasets/coco_detection.py

@@ -0,0 +1,50 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+
+dataset_type = 'CocoDataset'
+data_root = '/dataset/'
+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(
+    samples_per_gpu=2,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline))
+evaluation = dict(interval=1, metric='bbox')

detection/configs/_base_/datasets/coco_instance.py

@@ -0,0 +1,50 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+
+dataset_type = 'CocoDataset'
+data_root = '/dataset/'
+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(
+    samples_per_gpu=2,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline))
+evaluation = dict(metric=['bbox', 'segm'])

detection/configs/_base_/datasets/coco_instance_semantic.py

@@ -0,0 +1,56 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+
+dataset_type = 'CocoDataset'
+data_root = '/dataset/'
+
+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, with_seg=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='SegRescale', scale_factor=1 / 8),
+    dict(type='DefaultFormatBundle'),
+    dict(
+        type='Collect',
+        keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks', 'gt_semantic_seg']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(1333, 800),
+        flip=False,
+        transforms=[
+            dict(type='Resize', keep_ratio=True),
+            dict(type='RandomFlip', flip_ratio=0.5),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=2,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_train2017.json',
+        img_prefix=data_root + 'train2017/',
+        seg_prefix=data_root + 'stuffthingmaps/train2017/',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + 'annotations/instances_val2017.json',
+        img_prefix=data_root + 'val2017/',
+        pipeline=test_pipeline))
+evaluation = dict(metric=['bbox', 'segm'])

detection/configs/_base_/default_runtime.py

@@ -0,0 +1,17 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+
+checkpoint_config = dict(interval=1)
+# yapf:disable
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook')
+    ])
+# yapf:enable
+custom_hooks = [dict(type='NumClassCheckHook')]
+
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+load_from = None
+resume_from = None
+workflow = [('train', 1)]