8ac704ce-924d-4415-96d0-a7a53cd460d1

Build # 8ac704ce-924d-4415-96d0-a7a53cd460d1

Build Type

circle-ci

Committed by xzdandy

Commit Message

Merge branch 'obj-tracking' of github.com:georgia-tech-db/eva into obj-tracking

Pull Request Pull Request #566: feat: object tracking

Run Details

155 of 155 new or added lines in 16 files covered. (100.0%)

9371 of 9588 relevant lines covered (97.74%)

0.98 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

91.67

/eva/udfs/trackers/tracker.py

# coding=utf-8
# Copyright 2018-2022 EVA
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import typing

import numpy
import pandas as pd

from eva.catalog.catalog_type import NdArrayType
from eva.udfs.abstract.abstract_udf import AbstractUDF
from eva.udfs.decorators.decorators import forward, setup
from eva.udfs.decorators.io_descriptors.data_types import NumpyArray


class EVATracker(AbstractUDF):
    def __init__(self) -> None:
        pass

    def name(self):
        return "EVATracker"

    @setup(cachable=False, udf_type="object_tracker", batchable=False)
    def setup(self):
        pass

    @forward(
        input_signatures=[
            NumpyArray("frame_id", type=NdArrayType.INT32, dimensions=(1,)),
            NumpyArray("frame", type=NdArrayType.FLOAT32, dimensions=(None, None, 3)),
            NumpyArray("bboxes", type=NdArrayType.FLOAT32, dimensions=(None, 4)),
            NumpyArray("scores", type=NdArrayType.FLOAT32, dimensions=(None,)),
            NumpyArray("labels", type=NdArrayType.INT32, dimensions=(None,)),
        ],
        output_signatures=[
            NumpyArray("track_ids", type=NdArrayType.INT32, dimensions=(None,)),
            NumpyArray("track_labels", type=NdArrayType.INT32, dimensions=(None,)),
            NumpyArray("track_bboxes", type=NdArrayType.FLOAT32, dimensions=(None, 4)),
            NumpyArray("track_scores", type=NdArrayType.FLOAT32, dimensions=(None,)),
        ],
    )
    def forward(
        self,
        frame_id: numpy.ndarray,
        frame: numpy.ndarray,
        labels: numpy.ndarray,
        bboxes: numpy.ndarray,
        scores: numpy.ndarray,
    ) -> typing.Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, numpy.ndarray]:
        """
        Args:
            frame_id (numpy.ndarray): the frame id of current frame
            frame (numpy.ndarray): the input frame with shape (C, H, W)
            labels (numpy.ndarray): Corresponding labels for each box
            bboxes (numpy.ndarray): Array of shape `(n, 4)` or of shape `(4,)` where
            each row contains `(xmin, ymin, width, height)`.
            scores (numpy.ndarray): Corresponding scores for each box
        Returns:
            track_ids (numpy.ndarray): Corresponding track id for each box
            track_labels (numpy.ndarray): Corresponding labels for each box
            track_bboxes (numpy.ndarray):  Array of shape `(n, 4)` of tracked objects
            track_scores (numpy.ndarray): Corresponding scores for each box
        """
        raise NotImplementedError

    def __call__(self, *args, **kwargs):
        assert isinstance(
            args[0], pd.DataFrame
        ), f"Expecting pd.DataFrame, got {type(args[0])}"

        results = []
        for _, row in args[0].iterrows():
            tuple = (
                numpy.array(row[0]),
                numpy.array(row[1]),
                numpy.stack(row[2]),
                numpy.stack(row[3]),
                numpy.stack(row[4]),
            )
            results.append(self.forward(*tuple))
        return pd.DataFrame(
            results,
            columns=["track_ids", "track_labels", "track_bboxes", "track_scores"],
        )

1	# coding=utf-8
2	# Copyright 2018-2022 EVA
3	#
4	# Licensed under the Apache License, Version 2.0 (the "License");
5	# you may not use this file except in compliance with the License.
6	# You may obtain a copy of the License at
7	#
8	# http://www.apache.org/licenses/LICENSE-2.0
9	#
10	# Unless required by applicable law or agreed to in writing, software
11	# distributed under the License is distributed on an "AS IS" BASIS,
12	# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13	# See the License for the specific language governing permissions and
14	# limitations under the License.
15	import typing	1✔
16
17	import numpy	1✔
18	import pandas as pd	1✔
19
20	from eva.catalog.catalog_type import NdArrayType	1✔
21	from eva.udfs.abstract.abstract_udf import AbstractUDF	1✔
22	from eva.udfs.decorators.decorators import forward, setup	1✔
23	from eva.udfs.decorators.io_descriptors.data_types import NumpyArray	1✔
24
25
26	class EVATracker(AbstractUDF):	1✔
27	def __init__(self) -> None:	1✔
28	pass	1✔
29
30	def name(self):	1✔
31	return "EVATracker"	×
32
33	@setup(cachable=False, udf_type="object_tracker", batchable=False)	1✔
34	def setup(self):	1✔
35	pass	×
36
37	@forward(	1✔
38	input_signatures=[
39	NumpyArray("frame_id", type=NdArrayType.INT32, dimensions=(1,)),
40	NumpyArray("frame", type=NdArrayType.FLOAT32, dimensions=(None, None, 3)),
41	NumpyArray("bboxes", type=NdArrayType.FLOAT32, dimensions=(None, 4)),
42	NumpyArray("scores", type=NdArrayType.FLOAT32, dimensions=(None,)),
43	NumpyArray("labels", type=NdArrayType.INT32, dimensions=(None,)),
44	],
45	output_signatures=[
46	NumpyArray("track_ids", type=NdArrayType.INT32, dimensions=(None,)),
47	NumpyArray("track_labels", type=NdArrayType.INT32, dimensions=(None,)),
48	NumpyArray("track_bboxes", type=NdArrayType.FLOAT32, dimensions=(None, 4)),
49	NumpyArray("track_scores", type=NdArrayType.FLOAT32, dimensions=(None,)),
50	],
51	)
52	def forward(	1✔
53	self,
54	frame_id: numpy.ndarray,
55	frame: numpy.ndarray,
56	labels: numpy.ndarray,
57	bboxes: numpy.ndarray,
58	scores: numpy.ndarray,
59	) -> typing.Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, numpy.ndarray]:
60	"""
61	Args:
62	frame_id (numpy.ndarray): the frame id of current frame
63	frame (numpy.ndarray): the input frame with shape (C, H, W)
64	labels (numpy.ndarray): Corresponding labels for each box
65	bboxes (numpy.ndarray): Array of shape `(n, 4)` or of shape `(4,)` where
66	each row contains `(xmin, ymin, width, height)`.
67	scores (numpy.ndarray): Corresponding scores for each box
68	Returns:
69	track_ids (numpy.ndarray): Corresponding track id for each box
70	track_labels (numpy.ndarray): Corresponding labels for each box
71	track_bboxes (numpy.ndarray): Array of shape `(n, 4)` of tracked objects
72	track_scores (numpy.ndarray): Corresponding scores for each box
73	"""
74	raise NotImplementedError
75
76	def __call__(self, args, *kwargs):	1✔
77	assert isinstance(	1✔
78	args[0], pd.DataFrame
79	), f"Expecting pd.DataFrame, got {type(args[0])}"
80
81	results = []	1✔
82	for _, row in args[0].iterrows():	1✔
83	tuple = (	1✔
84	numpy.array(row[0]),
85	numpy.array(row[1]),
86	numpy.stack(row[2]),
87	numpy.stack(row[3]),
88	numpy.stack(row[4]),
89	)
90	results.append(self.forward(*tuple))	1✔
91	return pd.DataFrame(	1✔
92	results,
93	columns=["track_ids", "track_labels", "track_bboxes", "track_scores"],
94	)

georgia-tech-db / eva / 8ac704ce-924d-4415-96d0-a7a53cd460d1

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous