add caffe-style resnet results in README

README.md

@@ -31,9 +31,18 @@ This is an official pytorch implementation of [*Simple Baselines for Human Pose
 | 256x192_pose_resnet_152_d256d256d256 | 0.720 | 0.893 | 0.798 | 0.687 | 0.789 | 0.778 | 0.934 | 0.846 | 0.736 | 0.839 |
 | 384x288_pose_resnet_152_d256d256d256 | 0.743 | 0.896 | 0.811 | 0.705 | 0.816 | 0.797 | 0.937 | 0.858 | 0.751 | 0.863 |
 
+
+#### Results on Caffe-style ResNet
+| Arch | AP | Ap .5 | AP .75 | AP (M) | AP (L) | AR | AR .5 | AR .75 | AR (M) | AR (L) |
+|---|---|---|---|---|---|---|---|---|---|---|
+| 256x192_pose_resnet_50_caffe_d256d256d256 | 0.704 | 0.914 | 0.782 | 0.677 | 0.744 | 0.735 | 0.921 | 0.805 | 0.704 | 0.783 |
+| 256x192_pose_resnet_101_caffe_d256d256d256 | 0.720 | 0.915 | 0.803 | 0.693 | 0.764 | 0.753 | 0.928 | 0.821 | 0.720 | 0.802 |
+
+
 ### Note:
 - Flip test is used.
 - Person detector has person AP of 56.4 on COCO val2017 dataset.
+- Difference between PyTorch-style and Caffe-style ResNet is the position of stride=2 convolution
 
 ## Environment
 The code is developed using python 3.6 on Ubuntu 16.04. NVIDIA GPUs are needed. The code is developed and tested using 4 NVIDIA P100 GPU cards. Other platforms or GPU cards are not fully tested.

experiments/coco/resnet101/256x192_d256x3_adam_lr1e-3_caffe.yaml

@@ -0,0 +1,77 @@
+GPUS: '0'
+DATA_DIR: ''
+OUTPUT_DIR: 'output'
+LOG_DIR: 'log'
+WORKERS: 4
+PRINT_FREQ: 100
+
+DATASET:
+  DATASET: 'coco'
+  ROOT: 'data/coco/'
+  TEST_SET: 'val2017'
+  TRAIN_SET: 'train2017'
+  FLIP: true
+  ROT_FACTOR: 40
+  SCALE_FACTOR: 0.3
+MODEL:
+  NAME: 'pose_resnet'
+  PRETRAINED: 'models/pytorch/imagenet/resnet101-caffe.pth'
+  STYLE: 'caffe'
+  IMAGE_SIZE:
+  - 192
+  - 256
+  NUM_JOINTS: 17
+  EXTRA:
+    TARGET_TYPE: 'gaussian'
+    HEATMAP_SIZE:
+    - 48
+    - 64
+    SIGMA: 2
+    FINAL_CONV_KERNEL: 1
+    DECONV_WITH_BIAS: false
+    NUM_DECONV_LAYERS: 3
+    NUM_DECONV_FILTERS:
+    - 256
+    - 256
+    - 256
+    NUM_DECONV_KERNELS:
+    - 4
+    - 4
+    - 4
+    NUM_LAYERS: 101
+LOSS:
+  USE_TARGET_WEIGHT: true
+TRAIN:
+  BATCH_SIZE: 32
+  SHUFFLE: true
+  BEGIN_EPOCH: 0
+  END_EPOCH: 140
+  RESUME: false
+  OPTIMIZER: 'adam'
+  LR: 0.001
+  LR_FACTOR: 0.1
+  LR_STEP:
+  - 90
+  - 120
+  WD: 0.0001
+  GAMMA1: 0.99
+  GAMMA2: 0.0
+  MOMENTUM: 0.9
+  NESTEROV: false
+TEST:
+  BATCH_SIZE: 32
+  COCO_BBOX_FILE: 'data/coco/person_detection_results/COCO_val2017_detections_AP_H_56_person.json'
+  BBOX_THRE: 1.0
+  FLIP_TEST: false
+  IMAGE_THRE: 0.0
+  IN_VIS_THRE: 0.2
+  MODEL_FILE: ''
+  NMS_THRE: 1.0
+  OKS_THRE: 0.9
+  USE_GT_BBOX: true
+DEBUG:
+  DEBUG: true
+  SAVE_BATCH_IMAGES_GT: true
+  SAVE_BATCH_IMAGES_PRED: true
+  SAVE_HEATMAPS_GT: true
+  SAVE_HEATMAPS_PRED: true

experiments/coco/resnet50/256x192_d256x3_adam_lr1e-3_caffe.yaml

@@ -15,7 +15,7 @@ DATASET:
   SCALE_FACTOR: 0.3
 MODEL:
   NAME: 'pose_resnet'
-  PRETRAINED: 'models/pytorch/imagenet/resnet50-19c8e357.pth'
+  PRETRAINED: 'models/pytorch/imagenet/resnet50-caffe.pth'
   STYLE: 'caffe'
   IMAGE_SIZE:
   - 192