Source code for wbia.algo.detect.selectivesearch

# -*- coding: utf-8 -*-
"""
Interface to Selective Search object proposals.
"""
import logging
import utool as ut
import vtool as vt
import tempfile
import subprocess
import shlex
import os
from os.path import abspath, dirname, expanduser, join, exists  # NOQA
import numpy as np
import scipy.io

(print, rrr, profile) = ut.inject2(__name__, '[selective search]')
logger = logging.getLogger('wbia')

# SCRIPT_PATH = abspath(dirname(__file__))
SCRIPT_PATH = abspath(expanduser(join('~', 'code', 'selective_search_ijcv_with_python')))

if not ut.get_argflag('--no-selective-search'):
    try:
        assert exists(SCRIPT_PATH)
    except AssertionError:
        logger.info(
            'WARNING Failed to find selective_search_ijcv_with_python. '
            'Selective Search is unavailable'
        )
        # if ut.SUPER_STRICT:
        #     raise


VERBOSE_SS = ut.get_argflag('--verbdss') or ut.VERBOSE


[docs]def detect_gid_list(ibs, gid_list, downsample=True, verbose=VERBOSE_SS, **kwargs): """ Args: gid_list (list of int): the list of IBEIS image_rowids that need detection downsample (bool, optional): a flag to indicate if the original image sizes should be used; defaults to True True: ibs.get_image_detectpaths() is used False: ibs.get_image_paths() is used Kwargs (optional): refer to the Selective Search documentation for configuration settings Args: ibs (wbia.IBEISController): image analysis api gid_list (list of int): the list of IBEIS image_rowids that need detection downsample (bool, optional): a flag to indicate if the original image sizes should be used; defaults to True Kwargs: detector, config_filepath, weights_filepath, verbose Yields: tuple: (gid, gpath, result_list) CommandLine: python -m wbia.algo.detect.selectivesearch detect_gid_list --show Example: >>> # DISABLE_DOCTEST >>> from wbia.algo.detect.selectivesearch import * # NOQA >>> from wbia.core_images import LocalizerConfig >>> import wbia >>> ibs = wbia.opendb('testdb1') >>> gid_list = ibs.get_valid_gids() >>> config = {'matlab_command': 'selective_search', 'verbose': True} >>> downsample = False >>> results_list = detect_gid_list(ibs, gid_list, downsample, **config) >>> results_list = list(results_list) >>> print('result lens = %r' % (map(len, list(results_list)))) >>> print('result[0] = %r' % (len(list(results_list[0][2])))) >>> config = {'matlab_command': 'selective_search_rcnn', 'verbose': True} >>> downsample = False >>> results_list = detect_gid_list(ibs, gid_list, downsample, **config) >>> results_list = list(results_list) >>> print('result lens = %r' % (map(len, list(results_list)))) >>> print('result[0] = %r' % (len(list(results_list[0][2])))) >>> ut.quit_if_noshow() >>> import wbia.plottool as pt >>> ut.show_if_requested() Yields: results (list of dict) """ # Get new gpaths if downsampling if downsample: gpath_list = ibs.get_image_detectpaths(gid_list) neww_list = [vt.open_image_size(gpath)[0] for gpath in gpath_list] oldw_list = [oldw for (oldw, oldh) in ibs.get_image_sizes(gid_list)] downsample_list = [oldw / neww for oldw, neww in zip(oldw_list, neww_list)] orient_list = [1] * len(gid_list) else: gpath_list = ibs.get_image_paths(gid_list) downsample_list = [None] * len(gpath_list) orient_list = ibs.get_image_orientation(gid_list) # Run detection results_iter = detect(gpath_list, verbose=verbose, **kwargs) # Upscale the results _iter = zip(downsample_list, gid_list, orient_list, results_iter) for downsample, gid, orient, (gpath, result_list) in _iter: # Upscale the results back up to the original image size for result in result_list: if downsample is not None and downsample != 1.0: for key in ['xtl', 'ytl', 'width', 'height']: result[key] = int(result[key] * downsample) bbox = ( result['xtl'], result['ytl'], result['width'], result['height'], ) bbox_list = [bbox] bbox = bbox_list[0] result['xtl'], result['ytl'], result['width'], result['height'] = bbox yield (gid, gpath, result_list)
[docs]def detect(gpath_list, matlab_command='selective_search', verbose=VERBOSE_SS, **kwargs): """ Args: gpath_list (list of str): the list of image paths that need proposal candidates Kwargs (optional): refer to the Selective Search documentation for configuration settings Returns: iter """ # Form the MATLAB script command that processes images and write to # temporary results file. temp_file, temp_filepath = tempfile.mkstemp(suffix='.mat') os.close(temp_file) gpath_str = '{%s}' % (','.join(["'%s'" % (gpath,) for gpath in gpath_list])) matlab_command_str = "%s(%s, '%s')" % (matlab_command, gpath_str, temp_filepath) if verbose: logger.info('Calling: %s' % (matlab_command_str,)) # Execute command in MATLAB. bash_command = 'matlab -nojvm -r "try; %s; catch; exit; end; exit"' bash_str = bash_command % (matlab_command_str,) bash_list = shlex.split(bash_str) with open('/dev/null', 'w') as null: process_id = subprocess.Popen(bash_list, stdout=null, cwd=SCRIPT_PATH) process_return_code = process_id.wait() if process_return_code != 0: raise RuntimeError('Matlab selective search did not exit successfully') # Read the results and undo Matlab's 1-based indexing. boxes_list = list(scipy.io.loadmat(temp_filepath)['all_boxes'][0]) subtractor = np.array((1, 1, 0, 0))[np.newaxis, :] results_list = [boxes - subtractor for boxes in boxes_list] if len(results_list) != len(gpath_list): raise ValueError('Matlab selective search did not return valid data') # Remove temporary file, and return. os.remove(temp_filepath) # Pack results results_list_ = [] for result_list in results_list: result_list_ = [] for result in result_list: xtl = int(np.around(result[0])) ytl = int(np.around(result[1])) xbr = int(np.around(result[2])) ybr = int(np.around(result[3])) result_dict = { 'xtl': xtl, 'ytl': ytl, 'width': xbr - xtl, 'height': ybr - ytl, 'class': None, 'confidence': 1.0, } result_list_.append(result_dict) results_list_.append(result_list_) results_list = zip(gpath_list, results_list_) return results_list