Hugging Face's logo

Datasets:
sebnem-byte
/
drone-navigation-event-camera

Tasks:
Robotics
Modalities:
Text
Formats:
text
Size:
< 1K
Tags:
event-camera
drone-navigation
slam
ros
uav
vio
Libraries:
Datasets
License:
Dataset card Data Studio Files
xet
 Community
1
sebnem-byte commited on
Commit
d2829b7
·
verified ·
1 Parent(s): 998d545

Upload 15 files

Browse files
Files changed (16) hide show
  1. .gitattributes +13 -0
  2. DAVIS_346.yaml +51 -0
  3. dataset_readme.txt +115 -0
  4. illumination/30_lit.bag +3 -0
  5. illumination/60_lit.bag +3 -0
  6. illumination/dyn_HDR.bag +3 -0
  7. illumination/low_lit.bag +3 -0
  8. illumination/static_HDR.bag +3 -0
  9. motion/filtered/circle_filtered.bag +3 -0
  10. motion/filtered/dark_manual_filtered.bag +3 -0
  11. motion/line.bag +3 -0
  12. motion/square.bag +3 -0
  13. motion/turner.bag +3 -0
  14. motion/unfiltered/auto_circle_unfiltered.bag +3 -0
  15. motion/unfiltered/manual_long_unfiltered_x.bag +3 -0
  16. motion/unfiltered/manual_unfiltered_y.bag +3 -0
.gitattributes CHANGED
@@ -58,3 +58,16 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
61
+ illumination/30_lit.bag filter=lfs diff=lfs merge=lfs -text
62
+ illumination/60_lit.bag filter=lfs diff=lfs merge=lfs -text
63
+ illumination/dyn_HDR.bag filter=lfs diff=lfs merge=lfs -text
64
+ illumination/low_lit.bag filter=lfs diff=lfs merge=lfs -text
65
+ illumination/static_HDR.bag filter=lfs diff=lfs merge=lfs -text
66
+ motion/filtered/circle_filtered.bag filter=lfs diff=lfs merge=lfs -text
67
+ motion/filtered/dark_manual_filtered.bag filter=lfs diff=lfs merge=lfs -text
68
+ motion/line.bag filter=lfs diff=lfs merge=lfs -text
69
+ motion/square.bag filter=lfs diff=lfs merge=lfs -text
70
+ motion/turner.bag filter=lfs diff=lfs merge=lfs -text
71
+ motion/unfiltered/auto_circle_unfiltered.bag filter=lfs diff=lfs merge=lfs -text
72
+ motion/unfiltered/manual_long_unfiltered_x.bag filter=lfs diff=lfs merge=lfs -text
73
+ motion/unfiltered/manual_unfiltered_y.bag filter=lfs diff=lfs merge=lfs -text
DAVIS_346.yaml ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ label: "DAVIS-IJRR17"
2
+ id: a0652606b3d9fd6b62f5448a9e1304be
3
+ cameras:
4
+ - camera:
5
+ label: dvs
6
+ id: 0
7
+ line-delay-nanoseconds: 0
8
+ image_height: 260
9
+ image_width: 346
10
+ type: pinhole
11
+ intrinsics:
12
+ cols: 1
13
+ rows: 4
14
+ data: [254.63685262627598, 253.65483910180083, 168.997010720912, 121.01418574521121]
15
+ distortion:
16
+ type: radial-tangential
17
+ parameters:
18
+ cols: 1
19
+ rows: 4
20
+ data: [-0.3754670162580797, 0.1311302616170846, 0.0013733901749410204, -0.0015906674110824643]
21
+ T_B_C:
22
+ cols: 4
23
+ rows: 4
24
+ data: [0.999905117246, -0.0122374970355, 0.00632456886338, 0.00674696422488,
25
+ 0.0121780171243, 0.999882045134, 0.00935904469797, 0.0007279224709,
26
+ -0.00643835413146, -0.00928113597812, 0.99993620202, 0.0342573613538,
27
+ 0.0, 0.0, 0.0, 1.0]
28
+ - camera:
29
+ label: standard
30
+ id: 1
31
+ line-delay-nanoseconds: 0
32
+ image_height: 260
33
+ image_width: 346
34
+ type: pinhole
35
+ intrinsics:
36
+ cols: 1
37
+ rows: 4
38
+ data: [254.63685262627598, 253.65483910180083, 168.997010720912, 121.01418574521121]
39
+ distortion:
40
+ type: radial-tangential
41
+ parameters:
42
+ cols: 1
43
+ rows: 4
44
+ data: [-0.3754670162580797, 0.1311302616170846, 0.0013733901749410204, -0.0015906674110824643]
45
+ T_B_C:
46
+ cols: 4
47
+ rows: 4
48
+ data: [0.999905117246, -0.0122374970355, 0.00632456886338, 0.00674696422488,
49
+ 0.0121780171243, 0.999882045134, 0.00935904469797, 0.0007279224709,
50
+ -0.00643835413146, -0.00928113597812, 0.99993620202, 0.0342573613538,
51
+ 0.0, 0.0, 0.0, 1.0]
dataset_readme.txt ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Edged USLAM Dataset – Davis346 UAV Sequences
2
+
3
+ This dataset was collected using a DAVIS346 event camera mounted on a UAV platform.
4
+ The recordings include synchronized events, frames, and IMU data, along with ground-truth poses from a motion capture system (Vicon).
5
+
6
+ --------------------------------------------------
7
+ 1. Sensor Topics
8
+ --------------------------------------------------
9
+ - Event camera (DAVIS346):
10
+ • /dvs/events → asynchronous event stream
11
+ • /dvs/image_raw → grayscale intensity frames
12
+ • /dvs/imu → inertial measurements (accelerometer + gyroscope)
13
+
14
+ - Ground truth (Vicon):
15
+ • /mavros/vision_pose/pose (used in motion/line.bag, motion/square.bag, motion/turner.bag)
16
+ • /local_pose_vicon/pose (used in filtered/ and unfiltered/ sequences)
17
+
18
+ --------------------------------------------------
19
+ 2. Dataset Structure
20
+ --------------------------------------------------
21
+ - motion/
22
+ - line.bag
23
+ - square.bag
24
+ - turner.bag
25
+ → Ground truth: /mavros/vision_pose/pose
26
+
27
+ - filtered/
28
+ → IR-filtered lens attached to the DAVIS346
29
+ → Ground truth: /local_pose_vicon/pose
30
+
31
+ - unfiltered/
32
+ → Default (non-filtered) lens on the DAVIS346
33
+ → Ground truth: /local_pose_vicon/pose
34
+
35
+ --------------------------------------------------
36
+ 3. Notes
37
+ --------------------------------------------------
38
+ - All sequences are timestamp-synchronized across topics.
39
+ - The IR filter (filtered/) improves robustness under strong illumination but reduces sensitivity.
40
+ - The unfiltered/ data preserve the original DAVIS346 response, including IR wavelengths.
41
+ - Ground truth in both cases provides 6-DoF UAV pose in the Vicon frame.
42
+
43
+ --------------------------------------------------
44
+ 4. Usage
45
+ --------------------------------------------------
46
+ When training or evaluating algorithms, ensure that the correct ground-truth topic is selected depending on the sequence folder.
47
+
48
+
49
+ # DAVIS346 Calibration Parameters (DVS_ext.yaml Explained)
50
+
51
+ This file contains the intrinsic, distortion, and extrinsic calibration parameters
52
+ for the DAVIS346 event camera (DAVIS-IJRR17 profile).
53
+
54
+ --------------------------------------------------
55
+ 1. Camera Information
56
+ --------------------------------------------------
57
+ - Resolution: 346 x 260 pixels
58
+ - Model: Pinhole (ideal projection with distortion correction)
59
+
60
+ Cameras:
61
+ - dvs (event stream)
62
+ - standard (grayscale frames)
63
+ Both share identical calibration values.
64
+
65
+ --------------------------------------------------
66
+ 2. Intrinsics (fx, fy, cx, cy)
67
+ --------------------------------------------------
68
+ fx = 254.6368
69
+ fy = 253.6548
70
+ cx = 168.9970
71
+ cy = 121.0141
72
+
73
+ Meaning:
74
+ - fx, fy: focal lengths in pixel units
75
+ - cx, cy: principal point (image center)
76
+
77
+ --------------------------------------------------
78
+ 3. Distortion Parameters (k1, k2, p1, p2)
79
+ --------------------------------------------------
80
+ k1 = -0.3755
81
+ k2 = 0.1311
82
+ p1 = 0.00137
83
+ p2 = -0.00159
84
+
85
+ Meaning:
86
+ - k1, k2: radial distortion (barrel/pincushion)
87
+ - p1, p2: tangential distortion (lens misalignment)
88
+
89
+ --------------------------------------------------
90
+ 4. Extrinsics (T_B_C)
91
+ --------------------------------------------------
92
+ Transformation from Camera to Body (IMU) frame:
93
+
94
+ Rotation matrix (approx. identity):
95
+ [ 0.9999 -0.0122 0.0063 ]
96
+ [ 0.0122 0.9999 0.0093 ]
97
+ [ -0.0064 -0.0093 0.9999 ]
98
+
99
+ Translation vector:
100
+ x = 0.0067 m (6.7 mm)
101
+ y = 0.0007 m (0.7 mm)
102
+ z = 0.0343 m (3.4 cm)
103
+
104
+ Meaning:
105
+ - Camera is placed ~3.4 cm ahead of the IMU,
106
+ with negligible rotation offset.
107
+
108
+ --------------------------------------------------
109
+ Summary
110
+ --------------------------------------------------
111
+ - Intrinsics define the internal camera model.
112
+ - Distortion corrects lens effects.
113
+ - Extrinsics align camera with IMU frame.
114
+ """
115
+
illumination/30_lit.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b44169f02fff0cdf924d0b71686e23f4000412a59ba3434b6a220294cda1e681
3
+ size 862611527
illumination/60_lit.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97c3796f32a672470bf29ae2bc960adbc4d5aeed4fcfc45de04906f6e11973b2
3
+ size 1004375627
illumination/dyn_HDR.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c30b5a10cb872b731bd710b12bebdb1ade5e7f5b0c5b45b7ecdee0e48f0794f9
3
+ size 1016074678
illumination/low_lit.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db755f1bdf3e5fa84758e4dfe993a85273cfbfbc57b16ea3f2051358f31bc2d3
3
+ size 711753749
illumination/static_HDR.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9505057d189bce4fa633e10d6237ecccf331e8d43f5abf74f1bef13985b2cb0
3
+ size 881709997
motion/filtered/circle_filtered.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67c2b8b98b7b53986c3938c4c9a8c6132e723b5595988e99782e52ec72c858b1
3
+ size 517920567
motion/filtered/dark_manual_filtered.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:151279bb3502e514be4d9982e0620c975c51090f8a8d8e3b99211bb4e111f247
3
+ size 458541636
motion/line.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eb9d7fbbdd3f2cc495728cc867070989f84e187fe6db69f4d95e386d451b083
3
+ size 676815500
motion/square.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f59eeb84ba948fdf175a91462fe2e317668f37baa76a92cae261c105ec951ac7
3
+ size 955063996
motion/turner.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16284f028d89cb40cb297b45eef1eb4d7376c64b95b5ee61ef9db934546a6b75
3
+ size 1442980471
motion/unfiltered/auto_circle_unfiltered.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:705c8a23378d4de79f4fb6ed7cb1a23a37a514c8fcaf718f9a0b4976503f2602
3
+ size 715798807
motion/unfiltered/manual_long_unfiltered_x.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fa886cfab0bb6f2bfa74e65a8ebc0829e066ee7c32f1ee5ea58bb663916b74b
3
+ size 2044603964
motion/unfiltered/manual_unfiltered_y.bag ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d231a4234ea3fb92b627ef599330cb2933ce3b16b84f1e697f205eb655db33a
3
+ size 1218958398