def create_annotation(form, object_name, labels, frame): annotation = Region() annotation.object_name = object_name if form.cleaned_data['high_level']: annotation.full_frame = True annotation.x = 0 annotation.y = 0 annotation.h = 0 annotation.w = 0 else: annotation.full_frame = False annotation.x = form.cleaned_data['x'] annotation.y = form.cleaned_data['y'] annotation.h = form.cleaned_data['h'] annotation.w = form.cleaned_data['w'] annotation.text = form.cleaned_data['text'] annotation.metadata = form.cleaned_data['metadata'] annotation.frame = frame annotation.video = frame.video annotation.region_type = Region.ANNOTATION annotation.save() for lname in labels: if lname.strip(): dl, _ = Label.objects.get_or_create(name=lname, set="UI") rl = RegionLabel() rl.video = annotation.video rl.frame = annotation.frame rl.region = annotation rl.label = dl rl.save()
def perform_face_indexing(video_id): from dvaapp.models import Region,Frame,Video,IndexEntries from dvalib import indexer,detector from dvaapp.operations.video_processing import WFrame,WVideo from django.conf import settings from scipy import misc face_indexer = indexer.FacenetIndexer() dv = Video.objects.get(id=video_id) video = WVideo(dv, settings.MEDIA_ROOT) frames = Frame.objects.all().filter(video=dv) wframes = [WFrame(video=video, frame_index=df.frame_index, primary_key=df.pk) for df in frames] input_paths = {f.local_path(): f.primary_key for f in wframes} faces_dir = '{}/{}/detections'.format(settings.MEDIA_ROOT, video_id) indexes_dir = '{}/{}/indexes'.format(settings.MEDIA_ROOT, video_id) face_detector = detector.FaceDetector() aligned_paths = face_detector.detect(wframes) logging.info(len(aligned_paths)) faces = [] faces_to_pk = {} count = 0 for path, v in aligned_paths.iteritems(): for scaled_img, bb in v: d = Region() d.region_type = Region.DETECTION d.video = dv d.confidence = 100.0 d.frame_id = input_paths[path] d.object_name = "mtcnn_face" left, top, right, bottom = bb[0], bb[1], bb[2], bb[3] d.y = top d.x = left d.w = right - left d.h = bottom - top d.save() face_path = '{}/{}.jpg'.format(faces_dir, d.pk) output_filename = os.path.join(faces_dir, face_path) misc.imsave(output_filename, scaled_img) faces.append(face_path) faces_to_pk[face_path] = d.pk count += 1 dv.refresh_from_db() dv.detections = dv.detections + count dv.save() path_count, emb_array, entries, feat_fname, entries_fname = face_indexer.index_faces(faces, faces_to_pk, indexes_dir, video_id) i = IndexEntries() i.video = dv i.count = len(entries) i.contains_frames = False i.contains_detections = True i.detection_name = "Face" i.algorithm = 'facenet' i.entries_file_name = entries_fname.split('/')[-1] i.features_file_name = feat_fname.split('/')[-1] i.save()
def ssd_detect(video_id): """ This is a HACK since Tensorflow is absolutely atrocious in allocating and freeing up memory. Once a process / session is allocated a memory it cannot be forced to clear it up. As a result this code gets called via a subprocess which clears memory when it exits. :param video_id: :return: """ import django from PIL import Image sys.path.append(os.path.dirname(__file__)) os.environ.setdefault("DJANGO_SETTINGS_MODULE", "dva.settings") django.setup() from django.conf import settings from dvaapp.models import Video, Region, Frame from dvalib import entity, detector dv = Video.objects.get(id=video_id) frames = Frame.objects.all().filter(video=dv) v = entity.WVideo(dvideo=dv, media_dir=settings.MEDIA_ROOT) wframes = { df.pk: entity.WFrame(video=v, frame_index=df.frame_index, primary_key=df.pk) for df in frames } detection_count = 0 algorithm = detector.SSDetector() logging.info("starting detection {}".format(algorithm.name)) frame_detections = algorithm.detect(wframes.values()) for frame_pk, detections in frame_detections.iteritems(): for d in detections: dd = Region() dd.region_type = Region.DETECTION dd.video = dv dd.frame_id = frame_pk dd.object_name = d['name'] dd.confidence = d['confidence'] dd.x = d['left'] dd.y = d['top'] dd.w = d['right'] - d['left'] dd.h = d['bot'] - d['top'] dd.save() img = Image.open(wframes[frame_pk].local_path()) img2 = img.crop((d['left'], d['top'], d['right'], d['bot'])) img2.save("{}/{}/detections/{}.jpg".format(settings.MEDIA_ROOT, video_id, dd.pk)) detection_count += 1 dv.refresh_from_db() dv.detections = dv.detections + detection_count dv.save()
def annotate_entire_frame(request, frame_pk): frame = Frame.objects.get(pk=frame_pk) annotation = None if request.method == 'POST': if request.POST.get('text').strip() \ or request.POST.get('metadata').strip() \ or request.POST.get('object_name', None): annotation = Region() annotation.region_type = Region.ANNOTATION annotation.x = 0 annotation.y = 0 annotation.h = 0 annotation.w = 0 annotation.full_frame = True annotation.text = request.POST.get('text') annotation.metadata = request.POST.get('metadata') annotation.object_name = request.POST.get('object_name', 'frame_metadata') annotation.frame = frame annotation.video = frame.video annotation.save() for label_name in request.POST.get('tags').split(','): if label_name.strip(): if annotation: dl = RegionLabel() dl.video = frame.video dl.frame = frame dl.label = Label.objects.get_or_create(name=label_name, set="UI")[0] dl.region = annotation dl.save() else: dl = FrameLabel() dl.video = frame.video dl.frame = frame dl.label = Label.objects.get_or_create(name=label_name, set="UI")[0] dl.save() return redirect("frame_detail", pk=frame.pk)
def create_yolo_test_data(): import json import shutil import numpy as np import os from PIL import Image setup_django() from dvaapp.shared import handle_uploaded_file from django.core.files.uploadedfile import SimpleUploadedFile from dvaapp.models import Region, TEvent, Frame, AppliedLabel from dvaapp.tasks import extract_frames, export_video_by_id try: shutil.rmtree('tests/yolo_test') except: pass try: os.mkdir('tests/yolo_test') except: pass data = np.load('shared/underwater_data.npz') json_test = {} json_test['anchors'] = [(0.57273, 0.677385), (1.87446, 2.06253), (3.33843, 5.47434), (7.88282, 3.52778), (9.77052, 9.16828)] id_2_boxes = {} class_names = { 0: "red_buoy", 1: "green_buoy", 2: "yellow_buoy", 3: "path_marker", 4: "start_gate", 5: "channel" } for i, image in enumerate(data['images'][:500]): path = "tests/yolo_test/{}.jpg".format(i) Image.fromarray(image).save(path) id_2_boxes[path.split('/')[-1]] = data['boxes'][i].tolist() local('zip tests/yolo_test.zip -r tests/yolo_test/* ') fname = "tests/yolo_test.zip" name = "yolo_test" f = SimpleUploadedFile(fname, file(fname).read(), content_type="application/zip") dv = handle_uploaded_file(f, name) extract_frames(TEvent.objects.create(video=dv).pk) for df in Frame.objects.filter(video=dv): for box in id_2_boxes[df.name]: r = Region() r.video = dv r.frame = df c, top_x, top_y, bottom_x, bottom_y = box r.object_name = class_names[c] r.region_type = Region.ANNOTATION r.x = top_x r.y = top_y r.w = bottom_x - top_x r.h = bottom_y - top_y r.save() l = AppliedLabel() l.frame = df l.video = dv l.label_name = class_names[c] l.region = r l.save() export_video_by_id(TEvent.objects.create(video=dv).pk) try: shutil.rmtree('tests/yolo_test') except: pass
} labels = {k: Label.objects.create(name=v, set="test") for k, v in class_names} for i, image in enumerate(data['images'][:500]): path = "/Users/aub3/tests/yolo_test/{}.jpg".format(i) Image.fromarray(image).save(path) id_2_boxes[path.split('/')[-1]] = data['boxes'][i].tolist() local('zip /Users/aub3/tests/yolo_test.zip -r /Users/aub3/tests/yolo_test/* ') fname = "/Users/aub3/tests/yolo_test.zip" name = "yolo_test" f = SimpleUploadedFile(fname, file(fname).read(), content_type="application/zip") dv = handle_uploaded_file(f, name) perform_dataset_extraction(TEvent.objects.create(video=dv).pk) for df in Frame.objects.filter(video=dv): for box in id_2_boxes[df.name]: r = Region() r.video = dv r.frame = df c, top_x, top_y, bottom_x, bottom_y = box r.object_name = class_names[c] r.region_type = Region.ANNOTATION r.x = top_x r.y = top_y r.w = bottom_x - top_x r.h = bottom_y - top_y r.save() l = RegionLabel() l.frame = df l.video = dv l.label = labels[c] l.region = r l.save()
def create_yolo_test_data(): import json import shutil import numpy as np import os from PIL import Image setup_django() from dvaapp.shared import handle_uploaded_file from django.core.files.uploadedfile import SimpleUploadedFile from dvaapp.models import Region,TEvent,Frame, AppliedLabel from dvaapp.tasks import extract_frames,export_video_by_id try: shutil.rmtree('tests/yolo_test') except: pass try: os.mkdir('tests/yolo_test') except: pass data = np.load('shared/underwater_data.npz') json_test = {} json_test['anchors'] = [(0.57273, 0.677385), (1.87446, 2.06253), (3.33843, 5.47434), (7.88282, 3.52778), (9.77052, 9.16828)] id_2_boxes = {} class_names = { 0:"red_buoy", 1:"green_buoy", 2:"yellow_buoy", 3:"path_marker", 4:"start_gate", 5:"channel" } for i,image in enumerate(data['images'][:500]): path = "tests/yolo_test/{}.jpg".format(i) Image.fromarray(image).save(path) id_2_boxes[path.split('/')[-1]] = data['boxes'][i].tolist() local('zip tests/yolo_test.zip -r tests/yolo_test/* ') fname = "tests/yolo_test.zip" name = "yolo_test" f = SimpleUploadedFile(fname, file(fname).read(), content_type="application/zip") dv = handle_uploaded_file(f, name) extract_frames(TEvent.objects.create(video=dv).pk) for df in Frame.objects.filter(video=dv): for box in id_2_boxes[df.name]: r = Region() r.video = dv r.frame = df c , top_x, top_y, bottom_x, bottom_y = box r.object_name = class_names[c] r.region_type = Region.ANNOTATION r.x = top_x r.y = top_y r.w = bottom_x - top_x r.h = bottom_y - top_y r.save() l = AppliedLabel() l.frame = df l.video = dv l.label_name = class_names[c] l.region = r l.save() export_video_by_id(TEvent.objects.create(video=dv).pk) try: shutil.rmtree('tests/yolo_test') except: pass