Skip to content

Instantly share code, notes, and snippets.

@wootfish
Last active August 29, 2015 14:24
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save wootfish/8d98090ac2c2765af7a6 to your computer and use it in GitHub Desktop.
Save wootfish/8d98090ac2c2765af7a6 to your computer and use it in GitHub Desktop.
Demo of sine transform
# imports and basic notebook setup
from cStringIO import StringIO
import numpy as np
import scipy.ndimage as nd
import PIL.Image
from IPython.display import clear_output, Image, display
from google.protobuf import text_format
import os
import sys
import caffe
import itertools
def showarray(a, fmt='jpeg'):
a = np.uint8(np.clip(a, 0, 255))
f = StringIO()
PIL.Image.fromarray(a).save(f, fmt)
display(Image(data=f.getvalue()))
# a couple of utility functions for converting to and from Caffe's input image layout
def preprocess(net, img):
return np.float32(np.rollaxis(img, 2)[::-1]) - net.transformer.mean['data']
def deprocess(net, img):
return np.dstack((img + net.transformer.mean['data'])[::-1])
def make_step(net, step_size=1.5, end='inception_4c/output', jitter=32, clip=True):
'''Basic gradient ascent step.'''
src = net.blobs['data'] # input image is storred in Net's 'data' blob
dst = net.blobs[end]
ox, oy = np.random.randint(-jitter, jitter+1, 2)
src.data[0] = np.roll(np.roll(src.data[0], ox, -1), oy, -2) # apply jitter shift
net.forward(end=end)
dst.diff[:] = dst.data # specify the optimiation objective
net.backward(start=end)
g = src.diff[0]
# apply normaized ascent step to the input image
src.data[:] += step_size/np.abs(g).mean() * g
src.data[0] = np.roll(np.roll(src.data[0], -ox, -1), -oy, -2) # unshift image
if clip:
bias = net.transformer.mean['data']
src.data[:] = np.clip(src.data, -bias, 255-bias)
def deepdream(net, base_img, iter_n=10, octave_n=4, octave_scale=1.4, end='inception_4c/output', clip=True, **step_params):
# prepare base images for all octaves
octaves = [preprocess(net, base_img)]
for i in xrange(octave_n-1):
octaves.append(nd.zoom(octaves[-1], (1, 1.0/octave_scale,1.0/octave_scale), order=1))
src = net.blobs['data']
detail = np.zeros_like(octaves[-1]) # allocate image for network-produced details
for octave, octave_base in enumerate(octaves[::-1]):
h, w = octave_base.shape[-2:]
if octave > 0:
# upscale details from the previous octave
h1, w1 = detail.shape[-2:]
detail = nd.zoom(detail, (1, 1.0*h/h1,1.0*w/w1), order=1)
src.reshape(1,3,h,w) # resize the network's input image size
src.data[0] = octave_base+detail
for i in xrange(iter_n):
make_step(net, end=end, clip=clip, **step_params)
# visualization
vis = deprocess(net, src.data[0])
if not clip: # adjust image contrast if clipping is disabled
vis = vis*(255.0/np.percentile(vis, 99.98))
#showarray(vis)
print octave, i, end, vis.shape
clear_output(wait=True)
# extract details produced on the current octave
detail = src.data[0]-octave_base
# returning the resulting image
return deprocess(net, src.data[0])
model_path = '../caffe/models/bvlc_googlenet/'
net_fn = model_path + 'deploy.prototxt'
param_fn = model_path + 'bvlc_googlenet.caffemodel'
# Patching model to be able to compute gradients.
# Note that you can also manually add "force_backward: true" line to "deploy.prototxt".
model = caffe.io.caffe_pb2.NetParameter()
text_format.Merge(open(net_fn).read(), model)
model.force_backward = True
open('tmp.prototxt', 'w').write(str(model))
net = caffe.Classifier('tmp.prototxt', param_fn,
mean = np.float32([104.0, 116.0, 122.0]), # ImageNet mean, training set dependent
channel_swap = (2,1,0)) # the reference model has channels in BGR order instead of RGB
try:
inputfile = sys.argv[1]
except IndexError:
inputfile = "sky1024px.jpg"
framedir = inputfile.split(".")[0]+"_frames"
try:
os.mkdir(framedir)
except OSError: # assume errno=17 (directory already exists)
pass
print "Opening", inputfile
img = np.float32(PIL.Image.open(inputfile))
frame = img
h, w = frame.shape[:2]
s = 0.017
theta = 0.1
def shiftfunc(n):
return int(17 * np.sin(n/17.))/5
for i in itertools.count(0):
print "\n\n===Saving the results after %i rounds===" % (i,)
PIL.Image.fromarray(np.uint8(frame)).save(framedir+"/%04d.jpg"%i)
# since this is kind of a big shift, we only apply it every other iteration
if i % 2 == 0:
for n in range(frame.shape[0]):
frame[:, n] = np.roll(frame[:, n], 3*shiftfunc(n))
frame = deepdream(net, frame, iter_n=9, octave_scale=1.35)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment