Skip to content
View Haris-bin-shakeel's full-sized avatar
4 followers · 5 following

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Pull Shark

Highlights

  • Pro

Block or report Haris-bin-shakeel

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Haris-bin-shakeel/README.md 
haris@github ──────────────────────────────
  Role         ML Engineer · CV · Remote Sensing
  Status       Research Intern @ DLL Lab, NUST SINES
  Education    BS CS, FAST-NUCES

  Contact ──────────────────────────────
  Email        muhammadharis.tech4@gmail.com
  Portfolio    haris-bin-shakeel.netlify.app
  LinkedIn     in/haris-shakeel-aa1186330

  Open Source ──────────────────────────────
+ Merged PRs   5+  (opencv, fiftyone, cellfinder, brainrender, movement)
  In review    NumPy · OpenCV · Metaflow

Research : machine learning for Earth observation

  • SAR-guided multimodal super-resolution — role-based adapters (SAR geometry, land-cover semantics, cross-attention) injected into a frozen diffusion SR backbone via forward hooks, with a conformal-calibrated uncertainty head that flags hallucinated detail
  • Point-supervised semantic segmentation — LoveDA; approaching full-supervision mIoU from a handful of labeled pixels per class
  • Multimodal satellite fusion — multi-source Earth observation for land-cover segmentation

Open source : merged PRs across five projects

Project Merged
opencv/opencv VolumeType enum name clash — macOS Objective-C bindings
voxel51/fiftyone FFmpeg passthrough via **kwargs in make_frames_dataset()
voxel51/fiftyone Hex-string keys preserved in the mask_targets label-schema sanitizer
brainglobe/cellfinder Keyword-only arguments with a deprecation wrapper
brainglobe/brainrender Video-save path fix without os.chdir
neuroinformatics-unit/movement compute_path_deviation — a v0.17.0 release highlight

In review: NumPy #31523 · OpenCV #29199 · Netflix/metaflow #3240


Now — Google Summer of Code proposals to OpenCV & pgmpy · open to ML engineer / CV research engineer roles → LinkedIn · muhammadharis.tech4@gmail.com

Pinned Loading

  1. opensr-conformal-uq opensr-conformal-uq Public

    Bounding confident hallucinations in satellite super-resolution using conformal risk control, adapting NeurIPS-style uncertainty guarantees to multi-spectral Earth observation imagery.

    Jupyter Notebook 2

  2. point-supervised-rs-segmentation point-supervised-rs-segmentation Public

    Sparse supervision semantic segmentation for remote sensing using Partial Focal Cross Entropy Loss , achieving near full-supervision mIoU with only sparse point annotations.

    Jupyter Notebook

  3. sar-guided-rgbn-super-resolution sar-guided-rgbn-super-resolution Public

  4. YT-shorts-autopilot YT-shorts-autopilot Public

    utonomous 7-module AI pipeline for YouTube Shorts — trend scraping, AI scripting, voice synthesis, cinematic assembly, and auto-scheduling.

    Python