RuntimeError: [enforce fail at pybind_state.cc:1111] success. Error running net keypoint_rcnn

[ezreal1129]

i have same problem with #35 ,and my json_dataset.py as follow: ##############################################################

Copyright (c) 2018-present, Facebook, Inc.

All rights reserved.

#

This source code is licensed under the license found in the

LICENSE file in the root directory of this source tree.

##############################################################

from future import absolute_import from future import division from future import print_function from future import unicode_literals

import os import numpy as np import scipy.sparse import cPickle as pickle import copy from tqdm import tqdm import math

import utils.boxes as box_utils from utils.timer import Timer

COCO API

from pycocotools.coco import COCO from pycocotools import mask as COCOmask

from core.config import cfg from utils.general import static_vars

import logging logger = logging.getLogger(name)

IM_DIR = 'image_directory' ANN_FN = 'annotation_file'

Set to true if the ROIDB needs to be split into frames

SPLIT_INTO_FRAMES = 'split_into_frames'

Set to true if the frames need to be decoded from videos

FRAMES_FROM_VIDEO = 'frames_from_video'

Function to read from the weakly labeled outputs

COMPUTED_ANNOTATIONS_INFO = 'computed_annotations_info'

Optional annotation directory. Used to store additional stuff like for

jsons for posetrack evaluations

ANN_DN = 'annotation_directory' DATASETS = { 'posetrack_v1.0_train': { IM_DIR: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrack/', ANN_FN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/lists/PoseTrack/v1.0/posetrack_train.json', ANN_DN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrackV1.0_Annots_train_json/', }, 'posetrack_v1.0_val': { IM_DIR: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrack/', ANN_FN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/lists/PoseTrack/v1.0/posetrack_val.json', ANN_DN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrackV1.0_Annots_val_json', }, 'posetrack_v1.0_test': { IM_DIR: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrack/', ANN_FN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/lists/PoseTrack/v1.0/posetrack_test.json', ANN_DN: '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrackV1.0_Annots_test_json', }, }

Important conventions for ROIDB

frame_id: 1-indexed. The reader is 0-indexed, so I make the conversion in

utils/image.py

class JsonDataset(object): def init(self, name): assert name in DATASETS.keys(), 'Unknown dataset name' logger.debug('Creating: {}'.format(name)) self.name = name self.image_directory = DATASETS[name][IM_DIR] self.debug_timer = Timer() self.COCO = COCO(DATASETS[name][ANN_FN]) self.annotation_directory = DATASETS[name][ANN_DN] if ANN_DN in \ DATASETS[name] else ''

Set up dataset classes

    category_ids = self.COCO.getCatIds()
    categories = [c['name'] for c in self.COCO.loadCats(category_ids)]
    self.category_to_id_map = dict(zip(categories, category_ids))
    self.classes = ['__background__'] + categories
    self.num_classes = len(self.classes)
    self.json_category_id_to_contiguous_id = {
        v: i + 1 for i, v in enumerate(self.COCO.getCatIds())}
    self.contiguous_category_id_to_json_id = {
        v: k for k, v in self.json_category_id_to_contiguous_id.items()}
    self._init_keypoints(name=self.name)
    # Added by rgirdhar: Used in tracking to know which is head keypoints,
    # when using PCK distance to connect the boxes
    self.person_cat_info = self.COCO.loadCats([
        self.category_to_id_map['person']])[0]
    # Added by rgirdhar: Set true if the frames need to be read out of a
    # video file
    self.frames_from_video = DATASETS[name][FRAMES_FROM_VIDEO] if \
        FRAMES_FROM_VIDEO in DATASETS[name] else False
    self.annotations_info = DATASETS[name][COMPUTED_ANNOTATIONS_INFO] if \
        COMPUTED_ANNOTATIONS_INFO in DATASETS[name] else None
    if self.annotations_info is not None:
        self.annotations_info['clip_length'] = self.annotations_info[
            'clip_length']()

def get_roidb(
        self, gt=False, proposal_file=None, min_proposal_size=2,
        proposal_limit=-1, crowd_filter_thresh=0):
    assert gt is True or crowd_filter_thresh == 0, \
        'Crowd filter threshold must be 0 if ground-truth annotations ' \
        'are not included.'
    image_ids = self.COCO.getImgIds()
    image_ids.sort()
    roidb = copy.deepcopy(self.COCO.loadImgs(image_ids))
    if len(cfg.ROIDB_SUBSET) > 0:
        roidb = roidb[cfg.ROIDB_SUBSET[0]: cfg.ROIDB_SUBSET[1]]
        logger.warning('Using a roidb subset {}'.format(cfg.ROIDB_SUBSET))
    annots = []
    if SPLIT_INTO_FRAMES in DATASETS[self.name] and DATASETS[
            self.name][SPLIT_INTO_FRAMES]:
        roidb, annots = self._split_roidb_frames(roidb)
    for entry in roidb:
        self._prep_roidb_entry(entry)
    if gt:
        # Include ground-truth object annotations
        self.debug_timer.tic()
        for entry_id, entry in enumerate(roidb):
            self._add_gt_annotations(entry, entry_id, annots)
        logger.debug('_add_gt_annotations took {:.3f}s'.
                     format(self.debug_timer.toc(average=False)))
    if proposal_file is not None:
        # Include proposals from a file
        self.debug_timer.tic()
        self._add_proposals_from_file(
            roidb, proposal_file, min_proposal_size, proposal_limit,
            crowd_filter_thresh)
        logger.debug('_add_proposals_from_file took {:.3f}s'.
                     format(self.debug_timer.toc(average=False)))
    _add_class_assignments(roidb)
    return roidb

def _prep_roidb_entry(self, entry):
    # Reference back to the parent dataset
    entry['dataset'] = self
    # Make file_name an abs path
    entry['image'] = os.path.join(self.image_directory, entry['file_name'])
    entry['flipped'] = False
    entry['has_visible_keypoints'] = False
    # Empty placeholders
    entry['boxes'] = np.empty((0, 4), dtype=np.float32)
    entry['tracks'] = np.empty((0, 1), dtype=np.int32)
    # head boxes, if available (like in PoseTrack)
    entry['head_boxes'] = np.empty((0, 4), dtype=np.float32)
    entry['segms'] = []
    entry['gt_classes'] = np.empty((0), dtype=np.int32)
    entry['seg_areas'] = np.empty((0), dtype=np.float32)
    entry['gt_overlaps'] = scipy.sparse.csr_matrix(np.empty(
        (0, self.num_classes), dtype=np.float32))
    entry['is_crowd'] = np.empty((0), dtype=np.bool)
    # 'box_to_gt_ind_map': Shape is (#rois). Maps from each roi to the index
    # in the list of rois that satisfy np.where(entry['gt_classes'] > 0)
    entry['box_to_gt_ind_map'] = np.empty((0), dtype=np.int32)
    if self.keypoints is not None:
        entry['gt_keypoints'] = np.empty(
            (0, 3, self.num_keypoints), dtype=np.int32)
    # Remove unwanted fields if they exist
    for k in ['date_captured', 'url', 'license', 'file_name']:
        if k in entry:
            del entry[k]

def convert_raw_predictions_to_objs(self, annots, image_id):
    if len(annots['boxes']) == 0:
        return []
    objs = []
    N = annots['boxes'].shape[0]
    for i in range(N):
        obj = {}
        # COCO labels are in xywh format, but I make predictions in xyxy
        # Remove the score from box before converting
        obj['bbox'] = box_utils.xyxy_to_xywh(annots['boxes'][i][
            np.newaxis, :4]).reshape((-1,)).tolist()
        obj['num_keypoints'] = annots['poses'][i].shape[-1]
        assert(obj['num_keypoints'] == cfg.KRCNN.NUM_KEYPOINTS)
        obj['segmentation'] = []
        obj['area'] = obj['bbox'][-1] * obj['bbox'][-2]
        obj['iscrowd'] = False
        pose = annots['poses'][i][:3].transpose()
        pose[pose[:, -1] >= 2.0, -1] = 2
        pose[pose[:, -1] < 2.0, -1] = 0
        obj['keypoints'] = pose.reshape((-1)).tolist()
        obj['track_id'] = annots['tracks'][i]
        obj['image_id'] = image_id
        obj['category_id'] = 1  # person
        objs.append(obj)
    return objs

def _add_gt_annotations(self, entry, entry_id, annots):
    if len(annots) > 0:
        objs = self.convert_raw_predictions_to_objs(
            annots[entry_id], entry['id'])
    else:
        ann_ids = self.COCO.getAnnIds(imgIds=entry['id'], iscrowd=None)
        objs = self.COCO.loadAnns(ann_ids)
    # Sanitize bboxes -- some are invalid
    valid_objs = []
    valid_segms = []
    width = entry['width']
    height = entry['height']
    for obj in objs:
        # crowd regions are RLE encoded and stored as dicts
        if isinstance(obj['segmentation'], list):
            # Valid polygons have >= 3 points, so require >= 6 coordinates
            obj['segmentation'] = [
                p for p in obj['segmentation'] if len(p) >= 6
            ]
        if obj['area'] < cfg.TRAIN.GT_MIN_AREA:
            continue
        if 'ignore' in obj and obj['ignore'] == 1:
            continue
        # Convert form x1, y1, w, h to x1, y1, x2, y2
        x1 = obj['bbox'][0]
        y1 = obj['bbox'][1]
        x2 = x1 + np.maximum(0., obj['bbox'][2] - 1.)
        y2 = y1 + np.maximum(0., obj['bbox'][3] - 1.)
        x1, y1, x2, y2 = box_utils.clip_xyxy_to_image(
            x1, y1, x2, y2, height, width)
        # Require non-zero seg area and more than 1x1 box size
        if obj['area'] > 0 and x2 > x1 and y2 > y1:
            obj['clean_bbox'] = [x1, y1, x2, y2]
            valid_objs.append(obj)
            valid_segms.append(obj['segmentation'])
    num_valid_objs = len(valid_objs)

    boxes = np.zeros((num_valid_objs, 4), dtype=entry['boxes'].dtype)
    tracks = -np.ones((num_valid_objs, 1), dtype=entry['tracks'].dtype)
    head_boxes = -np.ones((num_valid_objs, 4),
                          dtype=entry['head_boxes'].dtype)
    gt_classes = np.zeros((num_valid_objs), dtype=entry['gt_classes'].dtype)
    gt_overlaps = np.zeros(
        (num_valid_objs, self.num_classes),
        dtype=entry['gt_overlaps'].dtype)
    seg_areas = np.zeros((num_valid_objs), dtype=entry['seg_areas'].dtype)
    is_crowd = np.zeros((num_valid_objs), dtype=entry['is_crowd'].dtype)
    box_to_gt_ind_map = np.zeros(
        (num_valid_objs), dtype=entry['box_to_gt_ind_map'].dtype)
    if self.keypoints is not None:
        gt_keypoints = np.zeros(
            (num_valid_objs, 3, self.num_keypoints),
            dtype=entry['gt_keypoints'].dtype)

    im_has_visible_keypoints = False
    for ix, obj in enumerate(valid_objs):
        cls = self.json_category_id_to_contiguous_id[obj['category_id']]
        boxes[ix, :] = obj['clean_bbox']
        if 'track_id' in obj:
            tracks[ix, 0] = obj['track_id']
        if 'head_box' in obj:
            # NOTE: This box has NOT BEEN CLEANED, and NOT BEEN converted
            # to (xmin, ymin, xmax, ymax). This is only here to be used
            # in MPII evaluations
            head_boxes[ix, :] = obj['head_box']
        gt_classes[ix] = cls
        seg_areas[ix] = obj['area']
        is_crowd[ix] = obj['iscrowd']
        box_to_gt_ind_map[ix] = ix
        if self.keypoints is not None:
            gt_keypoints[ix, :, :] = self._get_gt_keypoints(obj)
            if np.sum(gt_keypoints[ix, 2, :]) > 0:
                im_has_visible_keypoints = True
        if obj['iscrowd']:
            # Set overlap to -1 for all classes for crowd objects
            # so they will be excluded during training
            gt_overlaps[ix, :] = -1.0
        else:
            gt_overlaps[ix, cls] = 1.0
    entry['boxes'] = np.append(entry['boxes'], boxes, axis=0)
    entry['tracks'] = np.append(entry['tracks'], tracks, axis=0)
    entry['head_boxes'] = np.append(entry['head_boxes'], head_boxes, axis=0)
    entry['segms'].extend(valid_segms)
    # To match the original implementation:
    # entry['boxes'] = np.append(
    #     entry['boxes'], boxes.astype(np.int).astype(np.float), axis=0)
    entry['gt_classes'] = np.append(entry['gt_classes'], gt_classes)
    entry['seg_areas'] = np.append(entry['seg_areas'], seg_areas)
    entry['gt_overlaps'] = np.append(
        entry['gt_overlaps'].toarray(), gt_overlaps, axis=0)
    entry['gt_overlaps'] = scipy.sparse.csr_matrix(entry['gt_overlaps'])
    entry['is_crowd'] = np.append(entry['is_crowd'], is_crowd)
    entry['box_to_gt_ind_map'] = np.append(
        entry['box_to_gt_ind_map'], box_to_gt_ind_map)
    if self.keypoints is not None:
        entry['gt_keypoints'] = np.append(
            entry['gt_keypoints'], gt_keypoints, axis=0)
        entry['has_visible_keypoints'] = im_has_visible_keypoints

def _add_proposals_from_file(
        self, roidb, proposal_file, min_proposal_size, top_k, crowd_thresh):
    logger.info('Loading proposals from: {}'.format(proposal_file))
    with open(proposal_file, 'r') as f:
        proposals = pickle.load(f)
    id_field = 'indexes' if 'indexes' in proposals else 'ids'  # compat fix
    _sort_proposals(proposals, id_field)
    box_list = []
    for i, entry in enumerate(roidb):
        if i % 2500 == 0:
            logger.info(' {:d}/{:d}'.format(i + 1, len(roidb)))
        boxes = proposals['boxes'][i]
        # Sanity check that these boxes are for the correct image id
        assert entry['id'] == proposals[id_field][i]
        # Remove duplicate boxes and very small boxes and then take top k
        boxes = box_utils.clip_boxes_to_image(
            boxes, entry['height'], entry['width'])
        keep = box_utils.unique_boxes(boxes)
        boxes = boxes[keep, :]
        keep = box_utils.filter_small_boxes(boxes, min_proposal_size)
        boxes = boxes[keep, :]
        if top_k > 0:
            boxes = boxes[:top_k, :]
        box_list.append(boxes)
    _merge_proposal_boxes_into_roidb(roidb, box_list)
    if crowd_thresh > 0:
        _filter_crowd_proposals(roidb, crowd_thresh)

def _init_keypoints(self, name=''):
    self.keypoints = None
    self.keypoint_flip_map = None
    self.keypoints_to_id_map = None
    self.num_keypoints = 0
    # Thus far only the 'person' category has keypoints
    if 'person' in self.category_to_id_map:
        cat_info = self.COCO.loadCats([self.category_to_id_map['person']])
    else:
        return

    # Check if the annotations contain keypoint data or not
    if 'keypoints' in cat_info[0]:
        keypoints = cat_info[0]['keypoints']
        self.keypoints_to_id_map = dict(
            zip(keypoints, range(len(keypoints))))
        self.keypoints = keypoints
        self.num_keypoints = len(keypoints)
        if name.startswith('keypoints_coco'):
            self.keypoint_flip_map = {
                'left_eye': 'right_eye',
                'left_ear': 'right_ear',
                'left_shoulder': 'right_shoulder',
                'left_elbow': 'right_elbow',
                'left_wrist': 'right_wrist',
                'left_hip': 'right_hip',
                'left_knee': 'right_knee',
                'left_ankle': 'right_ankle'}
        else:
            self.keypoint_flip_map = {
                'left_shoulder': 'right_shoulder',
                'left_elbow': 'right_elbow',
                'left_wrist': 'right_wrist',
                'left_hip': 'right_hip',
                'left_knee': 'right_knee',
                'left_ankle': 'right_ankle'}

def _get_gt_keypoints(self, obj):
    if 'keypoints' not in obj:
        return None
    kp = np.array(obj['keypoints'])
    x = kp[0::3]  # 0-indexed x coordinates
    y = kp[1::3]  # 0-indexed y coordinates
    # 0: not labeled; 1: labeled, not inside mask;
    # 2: labeled and inside mask
    v = kp[2::3]
    num_keypoints = len(obj['keypoints']) / 3
    assert num_keypoints == self.num_keypoints
    gt_kps = np.ones((3, self.num_keypoints), dtype=np.int32)
    for i in range(self.num_keypoints):
        gt_kps[0, i] = x[i]
        gt_kps[1, i] = y[i]
        gt_kps[2, i] = v[i]
    return gt_kps

def _split_roidb_frames(self, roidb):
    # Config options
    clips_per_video = cfg.VIDEO.DEFAULT_CLIPS_PER_VIDEO
    clip_length = 1  # 1-frame clips
    if self.annotations_info is not None:
        clips_per_video = self.annotations_info['clips_per_video']
        clip_length = self.annotations_info['clip_length']
        entry_to_shard = _assign_shard_id_to_roidb(
            roidb, self.annotations_info['num_splits'],
            self.annotations_info['tot_vids'])

    # For each video in roidb, split into a entry per-frame
    new_roidb = []
    new_annots = []
    for entry_id, entry in enumerate(tqdm(roidb, desc='Splitting video->frames')):
        assert 'nframes' in entry, 'Video dataset must have nframes'
        # Get annotations, if possible
        annots = {}
        if self.annotations_info is not None:
            annots = _read_weak_annotations(
                entry_to_shard[entry_id],
                data_dir=self.annotations_info['data_dir'],
                det_file_name=self.annotations_info['det_file_name'])
            assert(len(annots['boxes']) == entry['nframes'])
        # roidb frame_ids are 1-indexed
        already_added = {}  # don't add same frame multiple times
        step_size = max(entry['nframes'] // clips_per_video, 1)
        for start_frame_id in range(1, entry['nframes'] + 2 - clip_length,
                                    step_size):
            for frame_id in range(start_frame_id, start_frame_id + clip_length):
                if frame_id in already_added:
                    continue
                new_entry = copy.deepcopy(entry)
                new_entry['frame_id'] = frame_id
                new_roidb.append(new_entry)
                if len(annots) != 0:
                    new_annots.append({
                        # frame_id is 1-indexed
                        'boxes': annots['boxes'][frame_id - 1],
                        'poses': annots['poses'][frame_id - 1],
                        'tracks': annots['tracks'][frame_id - 1],
                    })
                already_added[frame_id] = True
    logger.info('New roidb size {}'.format(len(new_roidb)))
    return new_roidb, new_annots

def _merge_proposal_boxes_into_roidb(roidb, box_list): assert len(box_list) == len(roidb) for i, entry in enumerate(roidb): boxes = box_list[i] num_boxes = boxes.shape[0] gt_overlaps = np.zeros( (num_boxes, entry['gt_overlaps'].shape[1]), dtype=entry['gt_overlaps'].dtype) box_to_gt_ind_map = -np.ones( (num_boxes), dtype=entry['box_to_gt_ind_map'].dtype)

    # Note: unlike in other places, here we intentionally include all gt
    # rois, even ones marked as crowd. Boxes that overlap with crowds will
    # be filtered out later (see: _filter_crowd_proposals).
    gt_inds = np.where(entry['gt_classes'] > 0)[0]
    if len(gt_inds) > 0:
        gt_boxes = entry['boxes'][gt_inds, :]
        gt_classes = entry['gt_classes'][gt_inds]
        proposal_to_gt_overlaps = box_utils.bbox_overlaps(
            boxes.astype(dtype=np.float32, copy=False),
            gt_boxes.astype(dtype=np.float32, copy=False))
        # Gt box that overlaps each input box the most
        # (ties are broken arbitrarily by class order)
        argmaxes = proposal_to_gt_overlaps.argmax(axis=1)
        # Amount of that overlap
        maxes = proposal_to_gt_overlaps.max(axis=1)
        # Those boxes with non-zero overlap with gt boxes
        I = np.where(maxes > 0)[0]
        # Record max overlaps with the class of the appropriate gt box
        gt_overlaps[I, gt_classes[argmaxes[I]]] = maxes[I]
        box_to_gt_ind_map[I] = gt_inds[argmaxes[I]]
    entry['boxes'] = np.append(
        entry['boxes'],
        boxes.astype(entry['boxes'].dtype, copy=False),
        axis=0)
    entry['gt_classes'] = np.append(
        entry['gt_classes'],
        np.zeros((num_boxes), dtype=entry['gt_classes'].dtype))
    entry['seg_areas'] = np.append(
        entry['seg_areas'],
        np.zeros((num_boxes), dtype=entry['seg_areas'].dtype))
    entry['gt_overlaps'] = np.append(
        entry['gt_overlaps'].toarray(), gt_overlaps, axis=0)
    entry['gt_overlaps'] = scipy.sparse.csr_matrix(entry['gt_overlaps'])
    entry['is_crowd'] = np.append(
        entry['is_crowd'],
        np.zeros((num_boxes), dtype=entry['is_crowd'].dtype))
    entry['box_to_gt_ind_map'] = np.append(
        entry['box_to_gt_ind_map'],
        box_to_gt_ind_map.astype(
            entry['box_to_gt_ind_map'].dtype, copy=False))

def _filter_crowd_proposals(roidb, crowd_thresh): """Finds proposals that are inside crowd regions and marks them as overlap = -1 with each ground-truth rois, which means they will be excluded from training. """ for entry in roidb: gt_overlaps = entry['gt_overlaps'].toarray() crowd_inds = np.where(entry['is_crowd'] == 1)[0] non_gt_inds = np.where(entry['gt_classes'] == 0)[0] if len(crowd_inds) == 0 or len(non_gt_inds) == 0: continue crowd_boxes = box_utils.xyxy_to_xywh(entry['boxes'][crowd_inds, :]) non_gt_boxes = box_utils.xyxy_to_xywh(entry['boxes'][non_gt_inds, :]) iscrowd_flags = [int(True)] * len(crowd_inds) ious = COCOmask.iou(non_gt_boxes, crowd_boxes, iscrowd_flags) bad_inds = np.where(ious.max(axis=1) > crowd_thresh)[0] gt_overlaps[non_gt_inds[bad_inds], :] = -1 entry['gt_overlaps'] = scipy.sparse.csr_matrix(gt_overlaps)

def _add_class_assignments(roidb): for entry in roidb: gt_overlaps = entry['gt_overlaps'].toarray()

max overlap with gt over classes (columns)

    max_overlaps = gt_overlaps.max(axis=1)
    # gt class that had the max overlap
    max_classes = gt_overlaps.argmax(axis=1)
    entry['max_classes'] = max_classes
    entry['max_overlaps'] = max_overlaps
    # sanity checks
    # if max overlap is 0, the class must be background (class 0)
    zero_inds = np.where(max_overlaps == 0)[0]
    assert all(max_classes[zero_inds] == 0)
    # if max overlap > 0, the class must be a fg class (not class 0)
    nonzero_inds = np.where(max_overlaps > 0)[0]
    assert all(max_classes[nonzero_inds] != 0)

def _sort_proposals(proposals, id_field): order = np.argsort(proposals[id_field]) fields_to_sort = ['boxes', id_field, 'scores'] for k in fields_to_sort: proposals[k] = [proposals[k][i] for i in order]

def add_proposals(roidb, rois, scales): """Add proposal boxes (rois) to an roidb that has ground-truth annotations but no proposals. If the proposals are not at the original image scale, specify the scale factor that separate them in scales. """ box_list = [] for i in range(len(roidb)): inv_im_scale = 1. / scales[i] idx = np.where(rois[:, 0] == i)[0] box_list.append(rois[idx, 1:] * inv_im_scale) _merge_proposal_boxes_into_roidb(roidb, box_list)

For historical consistency, not filter crowds (TODO(rbg): investigate)

# json_dataset._filter_crowd_proposals(roidb, cfg.TRAIN.CROWD_FILTER_THRESH)
_add_class_assignments(roidb)

def _assign_shard_id_to_roidb(roidb, num_splits, tot_vids): """ Returns: list with one element for each entry in roidb (shard_dir_name, (start_frame_id (0-indexed, included), end_frame_id (0-indexed, not included))) """ shards = [] vids_per_job = int(math.ceil(tot_vids / num_splits)) last_proc = 0 for start_id in range(num_splits): this_end_pos = min(last_proc + vids_per_job, tot_vids + 1) this_outdir = '{0:05d}range{1}_{2}'.format( start_id, last_proc, this_end_pos)

run through the entries that get assigned to this shard, and set

    # what frames out of it belong to which video.
    last_frame_proc = 0
    for i in range(last_proc, min(this_end_pos, len(roidb))):
        # start_id is included and last_proc is not, as happens in the
        # ROIDB_SUBSET code
        this_frame_proc = last_frame_proc + roidb[i]['nframes']
        shards.append((
            this_outdir, (last_frame_proc, this_frame_proc)))
        last_frame_proc = this_frame_proc
    last_proc = this_end_pos
return shards

def pickle_cached_load(fpath, cache): if fpath in cache: return cache[fpath] with open(fpath, 'r') as fin: data = pickle.load(fin) cache.clear() cache[fpath] = data return data

@static_vars(weak_annot_cache={}) def _read_weak_annotations(shard_info, data_dir='', det_file_name='detections.pkl', fixed_str='test/kinetics_unlabeled_train/keypoint_rcnn'): det_fpath = os.path.join(data_dir, shard_info[0], fixed_str, det_file_name) data = pickle_cached_load(det_fpath, _read_weak_annotations.weak_annot_cache) boxes = data['all_boxes'][1][shard_info[1][0]: shard_info[1][1]] poses = data['all_keyps'][1][shard_info[1][0]: shard_info[1][1]] tracks = data['all_tracks'][1][shard_info[1][0]: shard_info[1][1]] assert(len(boxes) == len(poses)) assert(len(boxes) == len(tracks)) return {'boxes': boxes, 'poses': poses, 'tracks': tracks}

the '/home/amax/Documents/sangyi/DetectAndTrack/lib/datasets/data/PoseTrack/' is my image directory.

how can i fix it, i need anyone's help! thank you very muchi.
@rohitgirdhar

[yanxiangyi]

Same with #34 . This is because the net is feed by no img data. Just following the README.md to use tools/gen_posetrack_json.py.

Your IM_DIR should be the directory to your renamed images. For me, it's:

'posetrack_v1.0_train': {
        IM_DIR: 'lib/datasets/data/PoseTrack/posetrack_data/images_renamed/',
        ANN_FN: 'lib/datasets/lists/PoseTrack/v1.0/posetrack_train.json',
        # ANN_DN: 'lib/datasets/data/PoseTrackV1.0_Annots_train_json/',
    }

So on and so forth.

[ezreal1129]

@yanxiangyi i fix it ,thank you very much

[my-hello-world]

@ezreal1129 @yanxiangyi can you help me? my issue as #62 thanks~