how do i get the specific emotion( the maximum value in the list of emotions)

[justinshenk]

how do i get the specific emotion( the maximum value in the list of emotions)

[JustinShenk]

Interesting, I don't know what might be causing this. It could be interesting to debug and see if it is indeed skipping frames.

…

On Thu, Sep 17, 2020 at 4:31 AM Gaurav Ajariwal ***@***.***> wrote: hi @JustinShenk <https://github.com/justinshenk> , thanks. I will do that. Also, there is one issue that I am facing. Not sure if you have faced or not, but sometimes i think the model fails to capture all the frames? For the video that I am processing, the frames should be 3000+ but I end up getting only 1500. Do you happen to know something about why that happens? Thanks in advance. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACOLMZELN4Q5I6ZH6WLKMRDSGFYIVANCNFSM4HPYGUKQ> .

[JustinShenk]

Hi @Philip-Leron, try version 19.0.0. I added detector.top_emotion(img), which outputs emotion, score of the first face found in the image.

[Philip-Leron]

Thanks Justin

…

Sent from my iPhone

On 27 May 2019, at 6:57 PM, Justin Shenk ***@***.***> wrote: Hi @Philip-Leron, try version 19.0.0. I added detector.top_emotion(img), which outputs emotion, score of the first face found in the image. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[JustinShenk]

Hi Gaurav, That functionality can be added to the Video class by reusing some code from top_emotion. Feel free to make a pull request.

…

On Fri, Sep 11, 2020 at 3:30 AM Gaurav Ajariwal ***@***.***> wrote: Hi, I am using this to extract emotions from a video file. I also need the top emotion, but I think your method top_emotion takes an image's input. Any suggestions how I should tackle this? I am using pretty much same code as yours from video-example.py — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACOLMZGM2J5SP7F5OMIV7Y3SFF4URANCNFSM4HPYGUKQ> .

[JustinShenk]

I am looking at the code for the first time in a while, and I see a capture_duration variable set to 1000 / fps. This is about 30 seconds for 30 fps. I can't answer the question why this is there right now since it is not documented, but is your video longer? I would first see if it is finishing early because of this mystery variable.

…

On Fri, Sep 18, 2020 at 4:37 PM Gaurav Ajariwal ***@***.***> wrote: @JustinShenk <https://github.com/justinshenk> I need to look up, but what I have noticed is that for your test.mp4 video, there should be 115 frames, which means 115 rows of 7 emotions, but the data.csv files output only 90. For my videos, the total frames are 3000, but the data.csv shows 1500. So it is probably either skipping frames, or the data writer function is not writing things correctly. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACOLMZBDO7CJOIP5IGHK2P3SGNWBNANCNFSM4HPYGUKQ> .

[vichidamba]

Is this error solved?

[Philip-Leron]

Yeah. İt was helpful during my final exam

…

On Fri, Nov 4, 2022, 3:40 AM vichidamba ***@***.***> wrote: Is this error solved? — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADUQOQNCXTJP2VNJPM7VPSLWGSAS7ANCNFSM4HPYGUKQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[vichidamba]

No, I am also getting a mismatch in the output frame count from data.csv. It seems the model is skipping predictions when a face is not detected in a frame instead of outputting zero at that frame. Any solution for this? @JustinShenk @Philip-Leron @0xinit

[Philip-Leron]

İ will have a look at that project again it is been long since i checked it out

…

On Fri, Nov 4, 2022, 4:55 PM vichidamba ***@***.***> wrote: No, I am also getting a mismatch in the output frame count from data.csv. It seems the model is skipping predictions when a face is to detected in a frame instead of outputting zero at that frame. Any solution for this? — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADUQOQIA6X3UUZMIXF4TGA3WGU5YFANCNFSM4HPYGUKQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>