update code

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2016 Yuwen Xiong
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

data/.gitignore

@@ -0,0 +1,8 @@
+selective_search*
+imagenet_models*
+fast_rcnn_models*
+faster_rcnn_models*
+rfcn_models*
+VOCdevkit*
+coco*
+cache

data/README.md

@@ -0,0 +1,69 @@
+This directory holds (*after you download them*):
+- Caffe models pre-trained on ImageNet
+- Faster R-CNN models
+- Symlinks to datasets
+
+To download Caffe models (ZF, VGG16) pre-trained on ImageNet, run:
+
+```
+./data/scripts/fetch_imagenet_models.sh
+```
+
+This script will populate `data/imagenet_models`.
+
+To download Faster R-CNN models trained on VOC 2007, run:
+
+```
+./data/scripts/fetch_faster_rcnn_models.sh
+```
+
+This script will populate `data/faster_rcnn_models`.
+
+In order to train and test with PASCAL VOC, you will need to establish symlinks.
+From the `data` directory (`cd data`):
+
+```
+# For VOC 2007
+ln -s /your/path/to/VOC2007/VOCdevkit VOCdevkit2007
+
+# For VOC 2012
+ln -s /your/path/to/VOC2012/VOCdevkit VOCdevkit2012
+```
+
+Install the MS COCO dataset at /path/to/coco
+
+```
+ln -s /path/to/coco coco
+```
+
+For COCO with Fast R-CNN, place object proposals under `coco_proposals` (inside
+the `data` directory). You can obtain proposals on COCO from Jan Hosang at
+https://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-computing/research/object-recognition-and-scene-understanding/how-good-are-detection-proposals-really/.
+For COCO, using MCG is recommended over selective search. MCG boxes can be downloaded
+from http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/.
+Use the tool `lib/datasets/tools/mcg_munge.py` to convert the downloaded MCG data
+into the same file layout as those from Jan Hosang.
+
+Since you'll likely be experimenting with multiple installs of Fast/er R-CNN in
+parallel, you'll probably want to keep all of this data in a shared place and
+use symlinks. On my system I create the following symlinks inside `data`:
+
+Annotations for the 5k image 'minival' subset of COCO val2014 that I like to use
+can be found at http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_minival2014.json.zip.
+Annotations for COCO val2014 (set) minus minival (~35k images) can be found at
+http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_valminusminival2014.json.zip.
+
+```
+# data/cache holds various outputs created by the datasets package
+ln -s /data/fast_rcnn_shared/cache
+
+# move the imagenet_models to shared location and symlink to them
+ln -s /data/fast_rcnn_shared/imagenet_models
+
+# move the selective search data to a shared location and symlink to them
+# (only applicable to Fast R-CNN training)
+ln -s /data/fast_rcnn_shared/selective_search_data
+
+ln -s /data/VOC2007/VOCdevkit VOCdevkit2007
+ln -s /data/VOC2012/VOCdevkit VOCdevkit2012
+```

data/pylintrc

@@ -0,0 +1,3 @@
+[TYPECHECK]
+
+ignored-modules = numpy, numpy.random, cv2

data/scripts/fetch_faster_rcnn_models.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=faster_rcnn_models.tgz
+URL=http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/$FILE
+CHECKSUM=ac116844f66aefe29587214272054668
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading Faster R-CNN demo models (695M)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."

data/scripts/fetch_imagenet_models.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=imagenet_models.tgz
+URL=http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/$FILE
+CHECKSUM=ed34ca912d6782edfb673a8c3a0bda6d
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading pretrained ImageNet models (1G)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."

data/scripts/fetch_selective_search_data.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=selective_search_data.tgz
+URL=http://www.cs.berkeley.edu/~rbg/fast-rcnn-data/$FILE
+CHECKSUM=7078c1db87a7851b31966b96774cd9b9
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading precomputed selective search boxes (0.5G)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."

experiments/README.md

@@ -0,0 +1,5 @@
+Scripts are under `experiments/scripts`.
+
+Each script saves a log file under `experiments/logs`.
+
+Configuration override files used in the experiments are stored in `experiments/cfgs`.

experiments/cfgs/faster_rcnn_alt_opt.yml

@@ -0,0 +1,5 @@
+EXP_DIR: faster_rcnn_alt_opt
+TRAIN:
+  BG_THRESH_LO: 0.0
+TEST:
+  HAS_RPN: True

experiments/cfgs/faster_rcnn_end2end.yml

@@ -0,0 +1,11 @@
+EXP_DIR: faster_rcnn_end2end
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.0
+TEST:
+  HAS_RPN: True

experiments/cfgs/rfcn_alt_opt_5step_ohem.yml

@@ -0,0 +1,12 @@
+EXP_DIR: rfcn_alt_opt_5step_ohem
+TRAIN:
+  BG_THRESH_LO: 0.0
+  RPN_PRE_NMS_TOP_N: 6000
+  RPN_POST_NMS_TOP_N: 300
+  AGNOSTIC: True
+  BATCH_SIZE: -1
+  RPN_NORMALIZE_TARGETS: True
+TEST:
+  PROPOSAL_METHOD: 'rpn'
+  HAS_RPN: False
+  AGNOSTIC: True

experiments/cfgs/rfcn_end2end.yml

@@ -0,0 +1,17 @@
+EXP_DIR: rfcn_end2end
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.1
+  BATCH_SIZE: 128
+  AGNOSTIC: True
+  SNAPSHOT_ITERS: 10000
+  RPN_PRE_NMS_TOP_N: 6000
+  RPN_POST_NMS_TOP_N: 300
+TEST:
+  HAS_RPN: True
+  AGNOSTIC: True

experiments/cfgs/rfcn_end2end_ohem.yml

@@ -0,0 +1,20 @@
+EXP_DIR: rfcn_end2end_ohem
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_NORMALIZE_TARGETS: True
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.0
+  BATCH_SIZE: -1
+  AGNOSTIC: True
+  SNAPSHOT_ITERS: 10000
+  RPN_PRE_NMS_TOP_N: 6000
+  RPN_POST_NMS_TOP_N: 300
+  ASPECT_GROUPING: False
+TEST:
+  HAS_RPN: True
+  AGNOSTIC: True
+  BBOX_REG: True

experiments/logs/.gitignore

@@ -0,0 +1 @@
+*.txt*

experiments/scripts/fast_rcnn.sh

@@ -0,0 +1,63 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/fast_rcnn.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/fast_rcnn.sh 0 VGG_CNN_M_1024 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_2007_trainval"
+    TEST_IMDB="voc_2007_test"
+    PT_DIR="pascal_voc"
+    ITERS=40000
+    ;;
+  coco)
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    ITERS=280000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/fast_rcnn_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/fast_rcnn/solver.prototxt \
+  --weights data/imagenet_models/${NET}.v2.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  ${EXTRA_ARGS}
+
+set +x
+NET_FINAL=`grep -B 1 "done solving" ${LOG} | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/fast_rcnn/test.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  ${EXTRA_ARGS}