helxsz

#download weights for vgg-f
wget https://www.dropbox.com/s/b96tfz2hnthfegm/vgg_f.npy

#download weights for caffenet
wget https://www.dropbox.com/s/8evo8ydl1jl03z4/caffenet.npy

#download weights for vgg-16
wget https://www.dropbox.com/s/zpeufcwesimhvua/vgg16.npy

#download weights for vgg-19

helxsz

# HED batch processing script; modified from https://github.com/s9xie/hed/blob/master/examples/hed/HED-tutorial.ipynb
# Step 1: download the hed repo: https://github.com/s9xie/hed
# Step 2: download the models and protoxt, and put them under {caffe_root}/examples/hed/
# Step 3: put this script under {caffe_root}/examples/hed/
# Step 4: run the following script: 
#       python batch_hed.py --images_dir=/data/to/path/photos/ --hed_mat_dir=/data/to/path/hed_mat_files/
# The code sometimes crashes after computation is done. Error looks like "Check failed: ... driver shutting down". You can just kill the job. 
# For large images, it will produce gpu memory issue. Therefore, you better resize the images before running this script. 
# Step 5: run the MATLAB post-processing script "PostprocessHED.m" 
# https://github.com/phillipi/pix2pix/blob/master/scripts/edges/batch_hed.py

helxsz

#!/usr/bin/env python

# --------------------------------------------------------
# Faster R-CNN
# Copyright (c) 2015 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ross Girshick
# --------------------------------------------------------

"""

helxsz

This document describes how to install the combination of 14.04 + CUDA 7.5 + Tensorflow. This combination is the easiest to install without anything like compilation from sources etc.

Download and install Ubuntu 14.04 or 16.04

http://releases.ubuntu.com/14.04/ http://releases.ubuntu.com/16.04/

Install CUDA 8

helxsz

# install dependencies
sudo apt-get update
sudo apt-get install -y build-essential
sudo apt-get install -y cmake
sudo apt-get install -y libgtk2.0-dev
sudo apt-get install -y pkg-config
sudo apt-get install -y python-numpy python-dev
sudo apt-get install -y libavcodec-dev libavformat-dev libswscale-dev
sudo apt-get install -y libjpeg-dev libpng12-dev libtiff5-dev libjasper-dev
 

helxsz

import caffe
import numpy as np
import cv2

def view_layer(net, layer_name, pad = 8, pad_color = 0.2):
    layer_data_raw = net.blobs[layer_name].data
    layer_data = layer_data_raw.copy()
    
    layer_data -= layer_data_raw.min()
    if layer_data_raw.max() - layer_data_raw.min() != 0:

helxsz

#coding=utf-8
import sys
import bpy
import math
import random
import colorsys
import os
import mathutils
import numpy as np
from random import randint

helxsz

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import matplotlib as mpl
from PIL import Image
import os


def read(file_path, image_path, vis=False, annotationPath=None):
	src_txt_dir = file_path

helxsz

# https://blender.stackexchange.com/questions/55965/creating-many-primitive-objects-quickly-with-python
import bpy, random
from random import randint

#Short hands for common calls
scene = bpy.context.scene
selected = bpy.context.selected_objects
mesh = bpy.ops.mesh
context = bpy.context
object = bpy.ops.object

helxsz

# https://cwlseu.github.io/receptive-field-arithmetic-for-CNN
# [filter size, stride, padding]
#Assume the two dimensions are the same
#Each kernel requires the following parameters:
# - k_i: kernel size
# - s_i: stride
# - p_i: padding (if padding is uneven, right padding will higher than left padding; "SAME" option in tensorflow)
# 
#Each layer i requires the following parameters to be fully represented: 
# - n_i: number of feature (data layer has n_1 = imagesize )