"Making Human-Robot Interaction Enjoyable"

Share and enjoy!

Are you also a Truman?

Posted by on Oct 27, 2017 in Press | 0 comments

Don Rowe wrote an excellent article about Jürgen’s and my work and ask the question, are you also just a Truman? Valid and important point. Have a read of the full article.

Talking to a robot.

Read More

Comparison of color measurement accuracy of ColorMunki Design and FRU WR-10QC Colorimeter

Posted by on Oct 25, 2017 in Design, Featured, Research | 0 comments

Comparison of color measurement accuracy of ColorMunki Design and FRU WR-10QC Colorimeter

I am working on a colour project and had purchased the WR10 colorimeter to complement my long serving work horse, the X-Rite Color Munki Design. My ColorMunki is already several years old and I was concerned that its accuracy might have declined. When I measured several hundreds of samples, I noticed that both colorimeters gave me considerably different LAB values.

To determine which device was closer to the truth I measured the 48 defined colours of Datacolor’s SpyderCHECKR 48. I calculated the absolute error both devices made. The results of a paired-sample t-test showed that the ColorMunki is producing significantly less measurement errors on L (t(47)=-9.229, p<0.001), L (t(47)=-4.590, p<0.001) and L (t(47)=-4.871, p<0.001). However, both devices measure colours that are significantly different from the target colour of the SpyderCheckr card on all three measurements. Figure 1 shows the means and standard deviation for all measurement errors.

Figure 1: Mean and Standard Deviation of all measurements for both devices.

There does seem to be some structure in the errors that WR-10 is producing. Have a look at the heat map (Figure 2). The data for my little experiment is available at the Open Science Framework (DOI: 10.17605/OSF.IO/UWEFD).

Figure 2: Heat Map of the absolute errors

Although both devices show some significant deviation from the original, it is not far off from what can be expected of devices in this price range. The ColorMunki Design produces significantly better results than the FRU’s WR-10QC.

Read More

Patch color definitions for Datacolor SpyderCheckr 48

Posted by on Oct 24, 2017 in Design, Documentation | 0 comments

Below you find the Datacolor SpyderCheckr 48 definition of patches in different color spaces, such as LAB, sRBG and AdobeRGB. Datacolor offers a lousy bitmap of the values which are difficult to read and impossible to use in a structured way. So there you go, a table of all the values that this color chart is suppose to represent:

Lab sRGB Adobe RGB
Number Patch Name L A B R G B R G B
1 1A Low Sat. Red 61.35 34.81 18.38 210 121 117 189 121 117
2 2A Low Sat. Yellow 75.5 5.84 50.42 216 179 90 205 178 96
3 3A Low Sat. Green 66.82 -25.1 23.47 127 175 120 141 174 122
4 4A Low Sat. Cyan 60.53 -22.6 -20.4 66 157 179 103 156 177
5 5A Low Sat. Blue 59.66 -2.03 -28.46 116 147 194 125 146 191
6 6A Low Sat. Magenta 59.15 30.83 -5.72 190 121 154 172 120 151
7 1B 10% Red Tint 82.68 5.03 3.02 218 203 201 213 202 200
8 2B 10% Green Tint 82.25 -2.42 3.78 203 205 196 202 204 195
9 3B 10% Blue Unit 82.29 2.2 -2.04 206 203 208 204 201 206
10 4B 90% Red Tone 24.89 4.43 0.78 66 57 58 66 60 60
11 5B 90% Green Tone 25.16 -3.88 2.13 54 61 56 59 63 59
12 6B 90% Blue Tone 26.13 2.61 -5.03 63 60 69 65 63 71
13 1C Lightest Skin 85.42 9.41 14.49 237 206 186 225 202 183
14 2C Lighter Skin 74.28 9.05 27.21 211 175 133 200 174 134
15 3C Moderate Skin 64.57 12.39 37.24 193 149 91 180 148 95
16 4C Medium Skin 44.49 17.23 26.24 139 93 61 127 93 65
17 SC Deep Skin 25.29 7.95 8.87 74 55 46 71 58 50
18 SC 95% Gray 22.67 2.11 -1.1 57 54 56 59 57 59
19 1D 5% Gray 92.72 1.89 2.76 241 233 229 238 233 229
20 2D 10% gray 88.85 1.59 2.27 229 222 220 226 221 219
21 3D 30% Gray 73.42 0.99 1.89 182 178 176 180 177 174
22 4D 50% Gray 57.15 0.57 1.19 139 136 135 137 135 134
23 5D 70% Gray 41.57 0.24 1.45 100 99 97 99 99 98
24 6D 90% Gray 25.65 1.24 0.05 63 61 62 65 63 64
25 1E Card White 96.04 2.16 2.6 249 242 238 247 242 237
26 2E 20% Gray 80.44 1.17 2.05 202 198 195 199 196 193
27 3E 40% Gray 65.52 0.69 1.86 161 157 154 158 156 153
28 4E 60% Gray 49.62 0.58 1.56 122 118 116 120 118 115
29 5E 80% Gray 33.55 0.35 1.4 80 80 78 81 81 79
30 6E Card Black 16.91 1.43 -0.81 43 41 43 46 46 47
31 1F Primary Cyan 47.12 -32.5 -28.75 0 127 159 39 126 157
32 2F Primary Magenta 50.49 53.45 -13.55 192 75 145 167 76 141
33 3F Primary Yellow 83.61 3.36 87.02 245 205 0 234 204 37
34 4F Primary Red 41.05 60.75 31.17 186 26 51 159 32 53
35 5F Primary Green 54.14 -40.8 34.75 57 146 64 94 145 71
36 6F Primary Blue 24.75 13.78 -49.48 25 55 135 41 58 132
37 1G Primary Orange 60.94 38.21 61.31 222 118 32 196 117 44
38 2G Blueprint 37.8 7.3 -43.04 99 86 96 70 89 156
39 3G Pink 49.81 48.5 15.76 195 79 95 170 80 94
40 4G Violet 28.88 19.36 -24.48 83 58 106 78 61 104
41 5G Apple Green 72.45 -23.6 60.47 157 188 54 165 186 69
42 6G Sunflower 71.65 23.74 72.28 236 158 25 218 157 46
43 1H Aqua 70.19 -31.9 1.98 98 187 166 130 186 166
44 2H Lavender 54.38 8.84 -25.71 126 125 174 125 124 171
45 3H Evergreen 42.03 -15.8 22.93 82 106 60 90 106 65
46 4H Steel Blue 48.82 -5.11 -23.08 87 120 155 98 119 152
47 SH Classic Light Skin 65.1 18.14 18.68 197 145 125 183 144 125
48 6H Classic Dark Skin 36.13 14.15 15.78 112 76 60 103 77 63
Read More

Insta360 Plugin For Adobe Premiere and After Effects Is Not Working

Posted by on Sep 28, 2017 in Design, Documentation | 1 comment

I recently recorded a lecture I gave in Stuttgart using the latest Insta360 One camera. Since it is very difficult to stop an academic from talking the whole presentation took nearly one hour. The camera divided the recoding into several files, each around 4 GB. This is probably because they wanted to be compatible with the old FAT32 file system.

Their 360 Editing Software allows you to convert their proprietary  INSV files into MP4 files and they even have a batch processing option. But the software cannot merge multiple videos into one, which I desperately needed.

Insta360 is also offering a plugin for Adobe Premiere and After Effects. After installing the plugin I was able to import the INSV files into Premiere, but the image of one of the cameras was upside down. I contacted their technical support and they explained to me that the plugin is currently not working.

Insta360 Adobe Plugin failure

I ended up having to convert all the INSV files to MP4 first and then editing them into one movie in Premiere. I have to admit that this workflow is rather inconvenient and I hope that Insta360 will either enable their editing software to merge movies or that they get their Adobe plugin working again. This seems another example of Banana Technology, it ripens at the user.

Read More

Persuasive Robotics Talk At The Emotional Machines Conference

Posted by on Sep 27, 2017 in Featured, Research | 0 comments

Persuasive Robotics Talk At The Emotional Machines Conference

I was invited to give a talk at the Interdisciplinary Conference on Emotional Machines in Stuttgart on September 21st, 2017. My talk focused mainly on the work I did in collaboration with Jürgen Brandstetter (doi: 10.1145/2909824.3020257, doi: 10.1177/0261927X15584682, doi: 10.1109/IROS.2014.6942730). My main argument was that the number of robots in our society will increase dramatically and robots will participate in the formation of our language. Through their influence on our language they will be able to nudge our valence related to certain terms. Moreover, it will only take 10% of us to own a robot for them to dominate the development of our language.

This is also the first time I used a 360 degree camera to record a talk. This technology becomes particularly useful when following the discussion between the speaker and the audience. YouTube’s 360 video feature does not work in all web browser (e.g. it does not work with Safari). Chrome and Firefox should be fine.

Read More

Affordable Sound Both for Voice Recording

Posted by on Sep 1, 2017 in Design, Documentation | 0 comments

Affordable Sound Both for Voice Recording

Recording a clear voice with no background noise is key for any video, audiobook or podcast. If you are limited in space and budget then you might consider building your own little sound proof voice recording box. I build this little gem from 6mm MDF and some acoustic foam taped to the box with double sided tape.

For the box you can create the drawing necessary for the laser cutter using online tools, such as Box Designer. It exports a PDF file which all the six faces clearly separated. If you are planning to make efficient use of your wood then you may prefer MakerCase. This service aligns the six sides so that they can be cut most efficiently from a piece of MDF. Here is the box drawing for laser cutter used for this design.

We attached rubber feed to the box and placed the microphone on an old mouse pad to isolate the microphone from any vibrations of the table, such as when a laptop is placed on it. Handles on the side allow for an easy transportation.

Here is a recording without the box. The microphone was placed just directly on the table:

This recording was made with the speaker sitting in front of the box speaking into the microphone.

The last example is with the speaker leaning into the box.

You will notice how all ambient noice is gone and how intimate the voice of the speaker sounds.

Read More

Bloomberg Businessweek interviewed Omics about my nonsense paper

Posted by on Aug 29, 2017 in Featured, Research | 0 comments

Bloomberg Businessweek interviewed Omics about my nonsense paper

Esmé E Deprez and Caroline Chen from Bloomberg Businessweek visited the headquarters of Omics in India to interview its owner Srinubabu Gedela about his company. Omics is widely considered a predatory publisher that publishes papers without rigorous peer review. Confronted with the acceptance of my non-sensical paper he replied that “Bartneck’s paper slipped through because it was submitted so close to the conference’s deadline.” Yeah, right.

Read More