From 117662d58786cade03b07493167f267220ba6b57 Mon Sep 17 00:00:00 2001 From: adamhrv Date: Sat, 6 Apr 2019 10:42:28 +0200 Subject: add duke mtmc saliency --- .../pages/datasets/brainwash/assets/background.jpg | Bin 78860 -> 100144 bytes .../brainwash/assets/brainwash_saliency_map.jpg | Bin 0 -> 46793 bytes .../pages/datasets/brainwash/assets/index.jpg | Bin 15644 -> 15565 bytes site/content/pages/datasets/brainwash/index.md | 4 +- .../pages/datasets/duke_mtmc/assets/background.jpg | Bin 205750 -> 149226 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam1.jpg | Bin 0 -> 76154 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam2.jpg | Bin 0 -> 69213 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam3.jpg | Bin 0 -> 99608 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam4.jpg | Bin 0 -> 102201 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam5.jpg | Bin 0 -> 65757 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam6.jpg | Bin 0 -> 110753 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam7.jpg | Bin 0 -> 125142 bytes .../duke_mtmc/assets/duke_mtmc_saliency_cam8.jpg | Bin 0 -> 85045 bytes .../pages/datasets/duke_mtmc/assets/index.jpg | Bin 22102 -> 17241 bytes site/content/pages/datasets/duke_mtmc/index.md | 89 ++++++++++++++++++++- .../pages/datasets/oxford_town_centre/index.md | 19 ++--- .../uccs/assets/.~lock.uccs_camera_exif.csv# | 1 - site/content/pages/datasets/uccs/index.md | 2 +- 18 files changed, 94 insertions(+), 21 deletions(-) mode change 100644 => 100755 site/content/pages/datasets/brainwash/assets/background.jpg create mode 100755 site/content/pages/datasets/brainwash/assets/brainwash_saliency_map.jpg mode change 100755 => 100644 site/content/pages/datasets/duke_mtmc/assets/background.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam1.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam2.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam3.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam4.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam5.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam6.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam7.jpg create mode 100644 site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam8.jpg mode change 100755 => 100644 site/content/pages/datasets/duke_mtmc/assets/index.jpg delete mode 100644 site/content/pages/datasets/uccs/assets/.~lock.uccs_camera_exif.csv# (limited to 'site/content/pages') diff --git a/site/content/pages/datasets/brainwash/assets/background.jpg b/site/content/pages/datasets/brainwash/assets/background.jpg old mode 100644 new mode 100755 index e6393ab5..83499088 Binary files a/site/content/pages/datasets/brainwash/assets/background.jpg and b/site/content/pages/datasets/brainwash/assets/background.jpg differ diff --git a/site/content/pages/datasets/brainwash/assets/brainwash_saliency_map.jpg b/site/content/pages/datasets/brainwash/assets/brainwash_saliency_map.jpg new file mode 100755 index 00000000..38e540f0 Binary files /dev/null and b/site/content/pages/datasets/brainwash/assets/brainwash_saliency_map.jpg differ diff --git a/site/content/pages/datasets/brainwash/assets/index.jpg b/site/content/pages/datasets/brainwash/assets/index.jpg index e5004ec0..d47aa68d 100755 Binary files a/site/content/pages/datasets/brainwash/assets/index.jpg and b/site/content/pages/datasets/brainwash/assets/index.jpg differ diff --git a/site/content/pages/datasets/brainwash/index.md b/site/content/pages/datasets/brainwash/index.md index 494555ab..1f13f568 100644 --- a/site/content/pages/datasets/brainwash/index.md +++ b/site/content/pages/datasets/brainwash/index.md @@ -25,12 +25,12 @@ Brainwash is not a widely used dataset but since its publication by Stanford Uni If you happen to have been at Brainwash cafe in San Francisco at any time on October 26, November 13, or November 24 in 2014 you are most likely included in the Brainwash dataset and have unwittingly contributed to surveillance research. -![caption: A visualization of 81,973 head annotations from the Brainwash dataset training partition. (c) Adam Harvey](assets/brainwash_mean_overlay.jpg) - {% include 'dashboard.html' %} {% include 'supplementary_header.html' %} +![caption: A visualization of 81,973 head annotations from the Brainwash dataset training partition. © megapixels.cc](assets/brainwash_saliency_map.jpg) + ![caption: An sample image from the Brainwash dataset used for training face and head detection algorithms for surveillance. The datset contains about 12,000 images. License: Open Data Commons Public Domain Dedication (PDDL)](assets/00425000_960.jpg) ![caption: 49 of the 11,918 images included in the Brainwash dataset. License: Open Data Commons Public Domain Dedication (PDDL)](assets/brainwash_montage.jpg) diff --git a/site/content/pages/datasets/duke_mtmc/assets/background.jpg b/site/content/pages/datasets/duke_mtmc/assets/background.jpg old mode 100755 new mode 100644 index 2275bd9e..080939ba Binary files a/site/content/pages/datasets/duke_mtmc/assets/background.jpg and b/site/content/pages/datasets/duke_mtmc/assets/background.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam1.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam1.jpg new file mode 100644 index 00000000..9d4388df Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam1.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam2.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam2.jpg new file mode 100644 index 00000000..2581ec33 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam2.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam3.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam3.jpg new file mode 100644 index 00000000..3d81c974 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam3.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam4.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam4.jpg new file mode 100644 index 00000000..e375abb9 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam4.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam5.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam5.jpg new file mode 100644 index 00000000..3d311f99 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam5.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam6.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam6.jpg new file mode 100644 index 00000000..f5b8d0d0 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam6.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam7.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam7.jpg new file mode 100644 index 00000000..116fe2ea Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam7.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam8.jpg b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam8.jpg new file mode 100644 index 00000000..f4b9e4e6 Binary files /dev/null and b/site/content/pages/datasets/duke_mtmc/assets/duke_mtmc_saliency_cam8.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/assets/index.jpg b/site/content/pages/datasets/duke_mtmc/assets/index.jpg old mode 100755 new mode 100644 index 7b144b42..1062e85c Binary files a/site/content/pages/datasets/duke_mtmc/assets/index.jpg and b/site/content/pages/datasets/duke_mtmc/assets/index.jpg differ diff --git a/site/content/pages/datasets/duke_mtmc/index.md b/site/content/pages/datasets/duke_mtmc/index.md index b8c359d2..e8c7556f 100644 --- a/site/content/pages/datasets/duke_mtmc/index.md +++ b/site/content/pages/datasets/duke_mtmc/index.md @@ -30,16 +30,97 @@ The 8 cameras deployed on Duke's campus were specifically setup to capture stude #### Data Visualizations -![caption: Duke MTMC pixel-averaged image of camera #5 is shown with the bounding boxes for each student drawn in white. (c) Adam Harvey](assets/duke_mtmc_cam5_average_comp.jpg) +=== columns 2 + +![caption: Camera 1 © megapixels.cc](assets/duke_mtmc_saliency_cam1.jpg) + +===== + +![caption: Camera 2 © megapixels.cc](assets/duke_mtmc_saliency_cam2.jpg) + +==== end columns + + +=== columns 2 + +![caption: Camera 3 © megapixels.cc](assets/duke_mtmc_saliency_cam3.jpg) + +===== + +![caption: Camera 4 © megapixels.cc](assets/duke_mtmc_saliency_cam4.jpg) + +==== end columns + + +=== columns 2 + +![caption: Camera 5 © megapixels.cc](assets/duke_mtmc_saliency_cam5.jpg) + +===== + +![caption: Camera 6 © megapixels.cc](assets/duke_mtmc_saliency_cam6.jpg) + +==== end columns + + +=== columns 2 + +![caption: Camera 7 © megapixels.cc](assets/duke_mtmc_saliency_cam7.jpg) + +===== + +![caption: Camera 8 © megapixels.cc](assets/duke_mtmc_saliency_cam8.jpg) + +==== end columns + + +### Alternate Layout + + +=== columns 4 + +![caption: Camera 1 © megapixels.cc](assets/duke_mtmc_saliency_cam1.jpg) + +===== + +![caption: Camera 2 © megapixels.cc](assets/duke_mtmc_saliency_cam2.jpg) + +==== + +![caption: Camera 3 © megapixels.cc](assets/duke_mtmc_saliency_cam3.jpg) + +==== + +![caption: Camera 4 © megapixels.cc](assets/duke_mtmc_saliency_cam4.jpg) + +==== end columns + + +=== columns 4 + +![caption: Camera 5 © megapixels.cc](assets/duke_mtmc_saliency_cam5.jpg) + +===== + +![caption: Camera 6 © megapixels.cc](assets/duke_mtmc_saliency_cam6.jpg) + +==== + +![caption: Camera 7 © megapixels.cc](assets/duke_mtmc_saliency_cam7.jpg) + +===== + +![caption: Camera 8 © megapixels.cc](assets/duke_mtmc_saliency_cam8.jpg) + +==== end columns + ### TODO -- change to heatmap overlay of each location -- make fancy viz of foot trails with bbox and blurred persons - expand story - add google street view images of each camera location? - add actual head detections to header image with faces blurred - add 4 diverse example images with faces blurred -- add map location of the brainwash cafe +- add links to google map locations of each camera ### Footnotes \ No newline at end of file diff --git a/site/content/pages/datasets/oxford_town_centre/index.md b/site/content/pages/datasets/oxford_town_centre/index.md index d2d37230..e65f27a9 100644 --- a/site/content/pages/datasets/oxford_town_centre/index.md +++ b/site/content/pages/datasets/oxford_town_centre/index.md @@ -3,7 +3,7 @@ status: published title: Oxford Town Centre desc: Oxford Town Centre is a dataset of surveillance camera footage from Cornmarket St Oxford, England -subdesc: The Oxford Town Centre dataset includes +subdesc: The Oxford Town Centre dataset includes approximately 2,200 identities and is used for research and development of face recognition systems slug: oxford_town_centre cssclass: dataset image: assets/background.jpg @@ -21,11 +21,9 @@ authors: Adam Harvey [ page under development ] -The Oxford Town Centre dataset is a video of pedestrians in a busy downtown area in Oxford used for creating surveillance algorithms with potential applications in activity recognition, remote biometric analysis, and non-cooperative face recognition. [^ben_benfold_orig] +The Oxford Town Centre dataset is a video of pedestrians in a busy downtown area in Oxford used for creating surveillance algorithms with potential applications in remote biometric analysis and non-cooperative face recognition.[^ben_benfold_orig] The dataset was originally created to build algorithms that improve the stability of pedestrian detectors to provide more accurate head location estimates, leading to more accurate face recognition. -REVISE - -Although Oxford Town Centre dataset first appears as a pedestrian dataset, it was created to improve the stabilization of pedstrian detections in order to extract a more accurate head region that would lead to improvements in face recognition. +Oxford Town Centre dataset is unique in that it uses footage from a public CCTV camera that is designated for public safety. Since its publication in 2009, the Oxford Town Centre CCTV footage dataset, and all 2,200 people in the video, have been redistributed around the world for the purpose of surveillance research and development. There are over 80 verified research projects that have used the Oxford Town Centre dataset. The usage even extends to commercial organizations including Amazon, Disney, and OSRAM. {% include 'dashboard.html' %} @@ -33,11 +31,9 @@ Although Oxford Town Centre dataset first appears as a pedestrian dataset, it wa ### Location -The street location of the camera used for the Oxford Town Centre dataset can be easily confirmed using only two visual clues in video: the GAP store and the main road [source](https://www.google.com/maps/@51.7528162,-1.2581152,3a,50.3y,310.59h,87.23t/data=!3m7!1e1!3m5!1s3FsGN-PqYC-VhQGjWgmBdQ!2e0!5s20120601T000000!7i13312!8i6656). The camera angle and field of view indicate that the camera was elevated and placed at the corner. The edge of the building is visible and there is a small white nylon strap and pigeon deterrent spikes visible on the upper perimeter of the building. Combined with stability of camera and pigeon appearances in front of the camera at 1:24 and 3:29, these visual cues indicate that the camera was mounted outside on the corner of the building just above the deterrence spikes. - -![caption: Footage from this public CCTV camera was used to create the Oxford Town Centre dataset. Image sources: Google Sreet View and Oxford Town Centre dataset](assets/oxford_town_centre_cctv.jpg) +The street location of the camera used for the Oxford Town Centre dataset can be easily confirmed using only two visual clues in video: the GAP store and the main road [source](https://www.google.com/maps/@51.7528162,-1.2581152,3a,50.3y,310.59h,87.23t/data=!3m7!1e1!3m5!1s3FsGN-PqYC-VhQGjWgmBdQ!2e0!5s20120601T000000!7i13312!8i6656). The camera angle and field of view indicate that the camera was elevated and placed at the corner. The edge of the building is visible and there is a small white nylon strap and pigeon deterrent spikes visible on the upper perimeter of the building. The field of view indicates the camera uses a wide angle lens. Combined with the camera's stability and pigeon appearances in front of the camera at 1:24 and 3:29, these visual cues indicate that the camera was mounted outside on the corner of the building just above the deterrence spikes. -Halfway through the video a peculiar and somewhat rude man enters the video and stands directly over top a water drain for over a minute. His unusual demeanor and apparently scripted behavior suggests a possible relationship to the CCTV operators. +![caption: Footage from this public CCTV camera was used to create the Oxford Town Centre dataset. Image sources: Google Street View and Oxford Town Centre dataset](assets/oxford_town_centre_cctv.jpg) ### Demo Videos Using Oxford Town Centre Dataset @@ -51,12 +47,9 @@ Several researchers have posted their demo videos using the Oxford Town Centre d - [VTD - towncenter.avi](https://www.youtube.com/watch?v=LwMOmqvhnoc) -[ add visualization ] - - TODO -- make visualization +- make heatmap viz - add license info diff --git a/site/content/pages/datasets/uccs/assets/.~lock.uccs_camera_exif.csv# b/site/content/pages/datasets/uccs/assets/.~lock.uccs_camera_exif.csv# deleted file mode 100644 index 61ef22ef..00000000 --- a/site/content/pages/datasets/uccs/assets/.~lock.uccs_camera_exif.csv# +++ /dev/null @@ -1 +0,0 @@ -,adam,adam,05.04.2019 12:39,file:///home/adam/.config/libreoffice/4; \ No newline at end of file diff --git a/site/content/pages/datasets/uccs/index.md b/site/content/pages/datasets/uccs/index.md index 8827b227..05e683af 100644 --- a/site/content/pages/datasets/uccs/index.md +++ b/site/content/pages/datasets/uccs/index.md @@ -80,7 +80,7 @@ The images in UCCS were taken on 18 non-consecutive days during 2012–2013. ### Location -The location of the camera and subjects can confirmed using the *Bellingcat method*. The visual clues that lead to location of the camera and subjects include the unique pattern of the sidewalk that is only used on the UCCS Pedestrian Spine near the West Lawn, the two UCCS sign poles with matching graphics still visible in Google Street View, the no parking sign and directionality of its arrow, the back of street sign next to it, the slight bend in the sidewalk, the presence of cars passing in the background of the image, and the far wall of the parking garage all match images in the dataset. The [original papers](https://www.semanticscholar.org/paper/Large-scale-unconstrained-open-set-face-database-Sapkota-Boult/07fcbae86f7a3ad3ea1cf95178459ee9eaf77cb1) also provides another clue: a [picture of the camera](https://www.semanticscholar.org/paper/Large-scale-unconstrained-open-set-face-database-Sapkota-Boult/07fcbae86f7a3ad3ea1cf95178459ee9eaf77cb1/figure/1) inside the office that was used to create the dataset. The window view in this image provides another match for the brick pattern on the north facade of the Kraember Family Library and the green metal fence along the sidewalk. View the [location on Google Maps](https://www.google.com/maps/place/University+of+Colorado+Colorado+Springs/@38.8934297,-104.7992445,27a,35y,258.51h,75.06t/data=!3m1!1e3!4m5!3m4!1s0x87134fa088fe399d:0x92cadf3962c058c4!8m2!3d38.8968312!4d-104.8049528) +The location of the camera and subjects can confirmed using several visual cues in the dataset images: the unique pattern of the sidewalk that is only used on the UCCS Pedestrian Spine near the West Lawn, the two UCCS sign poles with matching graphics still visible in Google Street View, the no parking sign and directionality of its arrow, the back of street sign next to it, the slight bend in the sidewalk, the presence of cars passing in the background of the image, and the far wall of the parking garage all match images in the dataset. The [original papers](https://www.semanticscholar.org/paper/Large-scale-unconstrained-open-set-face-database-Sapkota-Boult/07fcbae86f7a3ad3ea1cf95178459ee9eaf77cb1) also provides another clue: a [picture of the camera](https://www.semanticscholar.org/paper/Large-scale-unconstrained-open-set-face-database-Sapkota-Boult/07fcbae86f7a3ad3ea1cf95178459ee9eaf77cb1/figure/1) inside the office that was used to create the dataset. The window view in this image provides another match for the brick pattern on the north facade of the Kraember Family Library and the green metal fence along the sidewalk. View the [location on Google Maps](https://www.google.com/maps/place/University+of+Colorado+Colorado+Springs/@38.8934297,-104.7992445,27a,35y,258.51h,75.06t/data=!3m1!1e3!4m5!3m4!1s0x87134fa088fe399d:0x92cadf3962c058c4!8m2!3d38.8968312!4d-104.8049528) ![caption: Location on campus where students were unknowingly photographed with a telephoto lens to be used for defense and intelligence agency funded research on face recognition. Image: Google Maps](assets/uccs_map.jpg) -- cgit v1.2.3-70-g09d2