feat: add video samples to notebook

[bhavikapanara]

Added human fall video samples to the existing notebook. Here is the link to access the notebook. link

[commit-lint[bot]]

Features

• add plot title for video seq (db3d7c2)

Contributors

bhavikapanara

Commit-Lint commands

You can trigger Commit-Lint actions by commenting on this PR: - `@Commit-Lint merge patch` will merge dependabot PR on "patch" versions (X.X.Y - Y change) - `@Commit-Lint merge minor` will merge dependabot PR on "minor" versions (X.Y.Y - Y change) - `@Commit-Lint merge major` will merge dependabot PR on "major" versions (Y.Y.Y - Y change) - `@Commit-Lint merge disable` will desactivate merge dependabot PR - `@Commit-Lint review` will approve dependabot PR - `@Commit-Lint stop review` will stop approve dependabot PR

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[codecov[bot]]

Codecov Report

Merging #22 (db3d7c2) into main (b84a92d) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##             main      #22   +/-   ##
=======================================
  Coverage   92.29%   92.29%           
=======================================
  Files          14       14           
  Lines        1272     1272           
=======================================
  Hits         1174     1174           
  Misses         98       98           

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:13Z ----------------------------------------------------------------

bhavikapanara This is great, but it only shows the output of movenet. Please make this a comparison matrix with original, movenet and posenet as you did with the image samples.

_ivelin commented on 2021-09-08T18:38:40Z_ ----------------------------------------------------------------

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:14Z ----------------------------------------------------------------

bhavikapanara This is very helpful. It looks like part of the video at the end with the person actually fallen was cut off. The samples here only show the person before the fall. The last frame is with the person about to sit in a chair.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara same issue here. The actual fall frames aren't captured.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the fall frames.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:14Z ----------------------------------------------------------------

bhavikapanara This is very helpful. It looks like part of the video at the end with the person actually fallen was cut off. The samples here only show the person before the fall. The last frame is with the person about to sit in a chair.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara same issue here. The actual fall frames aren't captured.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the fall frames.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara same issue here. The actual fall frames aren't captured.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the fall frames.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z ----------------------------------------------------------------

bhavikapanara same issue here. The actual fall frames aren't captured.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the fall frames.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This sequence also misses the fall frames.

review-notebook-app[bot] commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:16Z ----------------------------------------------------------------

bhavikapanara This is a great example why we need to think more about the multi-person scenario. The models focus only on the person who has not fallen.

bhavikapanara commented 3 years ago

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

ivelin commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

bhavikapanara commented 3 years ago

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

ivelin commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

bhavikapanara commented 3 years ago

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

github-actions[bot] commented 2 years ago

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket:

• © Githubissues.
• Githubissues is a development platform for aggregating issues.

[bhavikapanara]

View / edit / reply to this conversation on ReviewNB

ivelin commented on 2021-09-02T16:27:15Z

bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames

[ivelin]

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes.

On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.***> wrote:

View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples.

Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png

— You are receiving this because your review was requested. Reply to this email directly, view it on GitHub https://github.com/ambianic/fall-detection/pull/22#issuecomment-912906764, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

[bhavikapanara]

I think it's reasonable to take snapshots once a second (1fps). That is how we implement it in production relying on the fact that falls take between 600-800ms. this should be sufficient to capture all before and after fall scenes. … On Fri, Sep 3, 2021 at 11:46 PM bhavika panara @.**> wrote: View / edit / reply to this conversation https://app.reviewnb.com/ambianic/fall-detection/pull/22/discussion/ on ReviewNB ivelin commented on 2021-09-02T16:27:15Z* bhavikapanara https://github.com/bhavikapanara This sequence also misses the frames with the person actually fallen on the floor. Can you check why that is the case and add them to the detection samples. Actually, I am skipping few frames because it is complicated to show all frames [image: image] https://user-images.githubusercontent.com/19815572/132082709-b2593118-2f06-49ef-95ed-40a33a829098.png — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AARBUFLOA6BMOE4ZF6OEG73UAGQD5ANCNFSM5DJAOSFQ .

@ivelin Please see the updated notebook

[ivelin]

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

--- View entire conversation on ReviewNB

[bhavikapanara]

@bhavikapanara Thank you for adding all important frames in the videos with before and after poses. Can you please print the confidence scores for keypoints under each detection frame as you did for the individual photos in this notebook.

Thank you!

View entire conversation on ReviewNB

@ivelin Please review the updated notebook

[github-actions[bot]]

:tada: This PR is included in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket: