# Request for Feedback – GSoC 2025 Proposal on OpenCV Model Zoo Quantization

**URL:** <https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683>\
**Category:** Uncategorized\
**Tags:** gsoc\
**Created:** [March 30, 2025, 10:30am UTC](https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683 "2025-03-30T10:30:52Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Devanshu\_Dasgupta](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/devanshu_dasgupta/32/12272_2.png) [@Devanshu\_Dasgupta](https://forum.opencv.org/u/Devanshu_Dasgupta)\
**Post date:** [March 30, 2025, 10:30am UTC](https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683/1 "2025-03-30T10:30:52Z")

</div>

Subject: Request for GSoC 2025 Proposal Feedback – OpenCV Model Zoo Quantization

Hello OpenCV Community,

I am Devanshu Dasgupta, an AI/ML enthusiast applying for Google Summer of Code (GSoC) 2025 under OpenCV. My project focuses on integrating quantized deep learning models into the OpenCV Model Zoo to enhance efficiency and edge deployment.

I have drafted my GSoC proposal and would love to get feedback from OpenCV mentors and contributors before submission.

Key Project Goals:  
✅ Implement Post-Training Quantization (PTQ) & Quantization-Aware Training (QAT) in OpenCV Model Zoo.  
✅ Optimize models for ARM, GPUs, and Edge TPUs.  
✅ Improve inference speed while maintaining accuracy.

📄 Proposal Link (Google Drive/GitHub Gist/Attachment): [Devanshu\_Dasgupta\_GSoC2025\_OpenCV\_Proposal-1.pdf - Google Drive](https://drive.google.com/file/d/16OmhfJBlmi7uDoNOD5ARt9XdwSZmxmXB/view?usp=drivesdk)

I would really appreciate any technical feedback, suggestions, or areas of improvement to strengthen my proposal.

Thank you for your time and guidance! Looking forward to your insights. 😊

Best Regards,  
Devanshu Dasgupta  
(Find me on LinkedIn: Devanshu Dasgupta)  
📧 Your Email: [devanshu.das.gupta14@gmail.com](mailto:devanshu.das.gupta14@gmail.com)

---

<div class="post-metadata">

**Author:** ![berak](https://avatars.discourse-cdn.com/v4/letter/b/85f322/32.png) [@berak](https://forum.opencv.org/u/berak)\
**Post date:** [March 30, 2025, 11:45am UTC](https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683/2 "2025-03-30T11:45:03Z")

</div>

you probably want to join [the mailing list](https://groups.google.com/g/opencv-gsoc-202x), too

---

<div class="post-metadata">

**Author:** ![Devanshu\_Dasgupta](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/devanshu_dasgupta/32/12272_2.png) [@Devanshu\_Dasgupta](https://forum.opencv.org/u/Devanshu_Dasgupta)\
**Post date:** [April 2, 2025, 8:54pm UTC](https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683/3 "2025-04-02T20:54:23Z")

</div>

Hello everyone, just following up on my GSoC proposal for ‘Quantized Models for OpenCV Model Zoo.’ I would appreciate any feedback from mentors or the community. Looking forward to your suggestions!

---

<div class="post-metadata">

**Author:** ![crackwitz](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/crackwitz/32/14_2.png) [@crackwitz](https://forum.opencv.org/u/crackwitz)\
**Post date:** [April 3, 2025, 10:15am UTC](https://forum.opencv.org/t/request-for-feedback-gsoc-2025-proposal-on-opencv-model-zoo-quantization/20683/4 "2025-04-03T10:15:29Z")

</div>

do not open a second thread about this for the sole purpose of “following up”. I have merged the new thread into this existing one.

did you read the response to your first thread (this one)?