- π₯ Multimodal Video Captioning - Audio-Visual understanding
- ποΈ Computer Vision - 3D Reconstruction, Pose Estimation
- π€ Vision Transformers - Attention mechanisms for visual tasks
- π Deep Learning Research - PyTorch implementations
- π Portfolio | π§ ashokbk215@gmail.com
-
03:57
(UTC +05:45) - github.com/blazewild
- https://www.asokbk.com.np/
- in/asokbk
Highlights
- Pro
Pinned Loading
-
Real-Time-Motion-Transfer-to-a-3D-Avatar
Real-Time-Motion-Transfer-to-a-3D-Avatar PublicReal-time human pose detection and motion transfer to 3D avatars using MediaPipe, DNN, and Three.js β supports webcam and video inputs with custom avatar integration.
-
FocalCap-Compressed-Video-Captioning
FocalCap-Compressed-Video-Captioning PublicAn extension of CoCap for fast and accurate compressed video captioning. FocalCap introduces Distilled Motion MAE pretraining and an AGDTR module to selectively enrich visual patches from H.264 encβ¦
C
-
MatchVision-AI-Sports-Video-Analytics-Tracking-Pipeline
MatchVision-AI-Sports-Video-Analytics-Tracking-Pipeline PublicMatchVision AI: A computer vision pipeline that converts broadcast football video into synchronized 2D tactical maps, tracking players, referees, and the ball using fine-tuned YOLO, ByteTrack, and β¦
Python
-
Custom_LLM_DataGen_Template
Custom_LLM_DataGen_Template Publicπ§ Modular pipeline for generating high-quality, domain-specific datasets for LLM fine-tuning β from PDFs and web scraping to synthetic Q&A generation, quality filtering, and training-ready formatting.
-
Blaze2Cap_AI_Motioner
Blaze2Cap_AI_Motioner Public3D Human Pose Estimation: BlazePose to TotalCapture Motion Dataset Pipeline with PyTorch DataLoader for motion capture research and machine learning
Python 2
-
TrekNepal-3B__Finetuned-Llama3.2-3B
TrekNepal-3B__Finetuned-Llama3.2-3B PublicFine-tuning pipeline for LLaMA 3.2-3B on Nepal trekking using custom synthetic Q&A data, LLM-based filtering, and QLoRA optimization.
Python
If the problem persists, check the GitHub status page or contact support.




