Update README.md #5
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Updated the readme to include log formats and new apk |
JackMostow
requested changes
Jul 27, 2021
There was a problem hiding this comment.
- Filename should include tablet ID (serial number).
- Filenames should include Unix epoch as well as human-readable timestamp.
- Good that you included an example! But it needs >1 face to disambiguate:
- It's not clear that "Bounding boxes" is list of strings and "HOGS" is a list of lists of floats.
- Please include an example to illustrate how they look when there are multiple faces.
- Specify explicitly if lists of bounding boxes, of landmarks, and of HOGS are in the same order.
- Specify whether faces are ordered by decreasing area.
- Specify that HOGs are omitted when resolution is too low to compute them.
|
Professor, the current version doesn’t have the the feature to accommodate
multiple faces for HOGs, it just takes the largest bounding box.
The unix epoch is there and I’ve mentioned it in the documentation
…On Tue, 27 Jul 2021 at 6:38 AM, JackMostow ***@***.***> wrote:
***@***.**** requested changes on this pull request.
- Filename should include tablet ID (serial number).
- Filenames should include Unix epoch as well as human-readable
timestamp.
- Good that you included an example! But it needs >1 face to
disambiguate:
- It's not clear that "Bounding boxes" is list of strings and "HOGS"
is a list of lists of floats.
- Please include an example to illustrate how they look when there are
multiple faces.
- Specify explicitly if lists of bounding boxes, of landmarks, and of
HOGS are in the same order.
- Specify whether faces are ordered by decreasing area.
- Specify that HOGs are omitted when resolution is too low to compute
them.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#5 (review)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMRYZCVLOMYOKVP5UQ4ZEP3TZYBIDANCNFSM5A7MOSQQ>
.
|
|
Can you make the syntax work for multiple faces in the future even though for now only one has HOGS?
From: NehaDalmia ***@***.***>
Sent: Tuesday, July 27, 2021 4:21 AM
To: RoboTutorLLC/RoboFeel ***@***.***>
Cc: Jack Mostow ***@***.***>; Mention ***@***.***>
Subject: Re: [RoboTutorLLC/RoboFeel] Update README.md (#5)
Professor, the current version doesn’t have the the feature to accommodate
multiple faces for HOGs, it just takes the largest bounding box.
The unix epoch is there and I’ve mentioned it in the documentation
On Tue, 27 Jul 2021 at 6:38 AM, JackMostow ***@***.***<mailto:***@***.***>> wrote:
***@***.**** requested changes on this pull request.
- Filename should include tablet ID (serial number).
- Filenames should include Unix epoch as well as human-readable
timestamp.
- Good that you included an example! But it needs >1 face to
disambiguate:
- It's not clear that "Bounding boxes" is list of strings and "HOGS"
is a list of lists of floats.
- Please include an example to illustrate how they look when there are
multiple faces.
- Specify explicitly if lists of bounding boxes, of landmarks, and of
HOGS are in the same order.
- Specify whether faces are ordered by decreasing area.
- Specify that HOGs are omitted when resolution is too low to compute
them.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#5 (review)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMRYZCVLOMYOKVP5UQ4ZEP3TZYBIDANCNFSM5A7MOSQQ>
.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub<#5 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AC64MWINUK64IOUCBDUJZXLTZZT55ANCNFSM5A7MOSQQ>.
|
|
Yes , it be changed in the future to accommodate multiple HOGs.
Even now it will have multiple bounding boxes with multiple faces, although
HOG is computed for the main face
…On Tue, Jul 27, 2021 at 7:31 PM JackMostow ***@***.***> wrote:
Can you make the syntax work for multiple faces in the future even though
for now only one has HOGS?
From: NehaDalmia ***@***.***>
Sent: Tuesday, July 27, 2021 4:21 AM
To: RoboTutorLLC/RoboFeel ***@***.***>
Cc: Jack Mostow ***@***.***>; Mention ***@***.***>
Subject: Re: [RoboTutorLLC/RoboFeel] Update README.md (#5)
Professor, the current version doesn’t have the the feature to accommodate
multiple faces for HOGs, it just takes the largest bounding box.
The unix epoch is there and I’ve mentioned it in the documentation
On Tue, 27 Jul 2021 at 6:38 AM, JackMostow ***@***.***<mailto:***@***.***>>
wrote:
> ***@***.**** requested changes on this pull request.
>
> - Filename should include tablet ID (serial number).
> - Filenames should include Unix epoch as well as human-readable
> timestamp.
> - Good that you included an example! But it needs >1 face to
> disambiguate:
> - It's not clear that "Bounding boxes" is list of strings and "HOGS"
> is a list of lists of floats.
> - Please include an example to illustrate how they look when there are
> multiple faces.
> - Specify explicitly if lists of bounding boxes, of landmarks, and of
> HOGS are in the same order.
> - Specify whether faces are ordered by decreasing area.
> - Specify that HOGs are omitted when resolution is too low to compute
> them.
>
> —
> You are receiving this because you authored the thread.
> Reply to this email directly, view it on GitHub
> <
#5 (review)>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AMRYZCVLOMYOKVP5UQ4ZEP3TZYBIDANCNFSM5A7MOSQQ>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub<
#5 (comment)>,
or unsubscribe<
https://github.com/notifications/unsubscribe-auth/AC64MWINUK64IOUCBDUJZXLTZZT55ANCNFSM5A7MOSQQ>.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMRYZCWJCZIWQOKDDMJID3DTZ232PANCNFSM5A7MOSQQ>
.
|
|
@NehaDalmia - Please clarify what "it" means in "it be changed in the future":
|
|
Professor what are you referring to with respect to “analysis code”, I’m
sorry I’m a bit confused
…On Wed, 28 Jul 2021 at 7:27 PM, JackMostow ***@***.***> wrote:
@NehaDalmia <https://github.com/NehaDalmia> - Please clarify what "it"
means in "it be changed in the future":
1. RoboFeel's ability to output HOGs for multiple faces? If so, good.
2. The ability of RoboFeel's output format to represent HOGs for
multliple faces? If so, not good, because it means that future analysis
code will require a different input format. I want the same analysis code
to handle RoboFeel output whether for 1 face or >1.
Thanks. - Jack
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMRYZCXTWUD47B27YLFNI7LT2AEEJANCNFSM5A7MOSQQ>
.
|
|
I mean whatever code is going to operate on your output files, e.g. to compute FAUs, infer affective states, correlate gaze with user actions (or inaction), etc. Initially I expect to do these operations off-line, but we'll eventually need to do them on-board if we want to use visual cues to guide RoboTutor behavior in realtime, not just inform its (re-)design. |
|
Oh sir, in that case I mean
1. RoboFeel's ability to output HOGs for multiple faces? If so, good.
I can clarify all this in the documentation
…On Wed, Jul 28, 2021 at 9:57 PM JackMostow ***@***.***> wrote:
I mean whatever code is going to operate on your output files, e.g. to
compute FAUs, infer affective states, correlate gaze with user actions (or
inaction), etc. Initially I expect to do these operations off-line, but
we'll eventually need to do them on-board if we want to use visual cues to
guide RoboTutor behavior in realtime, not just inform its (re-)design.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMRYZCTCSCGTO2ABGHVVMHLT2AVYBANCNFSM5A7MOSQQ>
.
|
|
Good. As I recall, when we discussed the file format, I had proposed grouping together all the information about each face (bounding box, landmarks, optional HOGs), but you explained that the standard representation groups information by type, and relies on listing faces in the same order for each type (which makes me nervous about inadvertently different orders). A consequence is that the list of HOGs will need to represent explicitly null HOGS for faces without them. Otherwise it may be ambiguous which faces have HOGs and which faces do not, unless we rely on risky assumptions about the order of faces in the representation. It's not an issue yet because for now only the first face will have HOG. But we should plan for multiple faces. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.