Hello Ajarn.......The first 50 secs, summed-up the scenario on chatgpt and prompt engineering.......With you, we never look at quick-fixes and only long term learning......Heard a lot about prompt engineering and this is the first online video, which I am watching.....And I am so glad, it has come from you and I will watch it till the end......
I missed hearing from you Mr. Kannan. I hope you are doing well. As you very well know, I rarely do videos on topics that are not directly related to data analysis, especially image analysis. But my day job is marketing and I benefit quite a bit from chat engines. Just trying to spread my knowledge and experience on the topic. I do worry if my audience can relate to such videos but the number of views and engagement gives me good feedback.
Oh man, i hope this is a one off stint. Really enjoyed your science-oriented videos. Have you covered volumetric measurements and analysis of z-stack of confocal/EM images?
I do these type of one-off videos during the Holiday time on topics of my personal interest and experience. They are still related as you will find data analysis and also a bit of coding. I have done segmentation of 3D objects in z-stack in the past: e.g., th-cam.com/video/-9w0oIpMgVw/w-d-xo.html I can explore the topic of 3D measurements if viewers are interested. Based on TH-cam video stats, my 3D videos are not doing as well as other 2D topics so I interpreted that as not many people being interested in 3D topics.
Hello sir! I am working on fish freshness image classification. I have tried using segmented images as input and passing them to a transfer learning based model. The dataset size is also large but still not getting good accuracy, I need you guidance sir.
I apologize for the inconvenience, but I am having trouble separating my crop fields. I am currently using SAM, but the fields are too similar in color. It is not working. Can you recommend a tool?
Why SAM? You challenge sounds like a semantic segmentation problem. Even if it isn't, SAM would not work right out of the box unless you train it further to customize for your work. I recommend using either U-net for Semantic or Mask R-CNN for instance. You can easily access Mask R-CNN via the Detectron2 library. Checkout my videos on Detectron2 topic where I provide a walkthrough from labeling to training to segmentation to measurements.
@@DigitalSreeni Thank you for your advice,Professor. I am a beginner. I am using segment-geospatial because I need to segment images with geographic information for use in QGIS. I will explore the Detectron2 library with your video tutorial. Your expertise is essential for my challenge of segmenting agricultural areas.
Dear stranger, please don't offer Sreeni 300k for prompt engineering. We do not need another one "elite cognac review with victoria secrets models" channel.
I appreciate the friendly advice! Rest assured, no one is going to pay me 300K to be a prompt engineer :) No fashion shows, just code shows on my channel. Thanks for looking out!
![[LIVE] : ONE ลุมพินี 84 วันนี้!! คู่เอก "ก้องศึก vs เมืองไทย"](http://i.ytimg.com/vi/3kuzsAIDf1c/mqdefault.jpg)
Hello Ajarn.......The first 50 secs, summed-up the scenario on chatgpt and prompt engineering.......With you, we never look at quick-fixes and only long term learning......Heard a lot about prompt engineering and this is the first online video, which I am watching.....And I am so glad, it has come from you and I will watch it till the end......
I missed hearing from you Mr. Kannan. I hope you are doing well.
As you very well know, I rarely do videos on topics that are not directly related to data analysis, especially image analysis. But my day job is marketing and I benefit quite a bit from chat engines. Just trying to spread my knowledge and experience on the topic. I do worry if my audience can relate to such videos but the number of views and engagement gives me good feedback.
You are excellent Sreeni
Always doing great job....Thanks srini
My pleasure 😊
Oh man, i hope this is a one off stint. Really enjoyed your science-oriented videos.
Have you covered volumetric measurements and analysis of z-stack of confocal/EM images?
I do these type of one-off videos during the Holiday time on topics of my personal interest and experience. They are still related as you will find data analysis and also a bit of coding.
I have done segmentation of 3D objects in z-stack in the past: e.g., th-cam.com/video/-9w0oIpMgVw/w-d-xo.html
I can explore the topic of 3D measurements if viewers are interested. Based on TH-cam video stats, my 3D videos are not doing as well as other 2D topics so I interpreted that as not many people being interested in 3D topics.
Hello, can you please post a video on binary segmentation using unet 3+
Hello sir! I am working on fish freshness image classification. I have tried using segmented images as input and passing them to a transfer learning based model. The dataset size is also large but still not getting good accuracy, I need you guidance sir.
I apologize for the inconvenience, but I am having trouble separating my crop fields. I am currently using SAM, but the fields are too similar in color. It is not working. Can you recommend a tool?
Why SAM? You challenge sounds like a semantic segmentation problem. Even if it isn't, SAM would not work right out of the box unless you train it further to customize for your work. I recommend using either U-net for Semantic or Mask R-CNN for instance. You can easily access Mask R-CNN via the Detectron2 library. Checkout my videos on Detectron2 topic where I provide a walkthrough from labeling to training to segmentation to measurements.
@@DigitalSreeni Thank you for your advice,Professor.
I am a beginner. I am using segment-geospatial because I need to segment images with geographic information for use in QGIS.
I will explore the Detectron2 library with your video tutorial. Your expertise is essential for my challenge of segmenting agricultural areas.
Dear stranger, please don't offer Sreeni 300k for prompt engineering. We do not need another one "elite cognac review with victoria secrets models" channel.
I appreciate the friendly advice! Rest assured, no one is going to pay me 300K to be a prompt engineer :) No fashion shows, just code shows on my channel. Thanks for looking out!