All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.

Deliver ultra-low bitrate image compression using a shallow encoder for efficient, real-time encoding on bandwidth and resource-limited devices.

🤖 Enhance programming education by fine-tuning the Phi-3 Mini model to deliver well-structured, documented code responses, ensuring best practices in coding.

[CVPR'26] Official Code for “V²-SAM: Marrying SAM2 with Multi-Prompt Experts for Cross-View Object Correspondence”

SimpleAICV:pytorch training examples.

Pytorch implementation of "EdgeCrafter: Compact ViTs for Edge Dense Prediction via Task-Specialized Distillation"

Scripts and documentation for converting the Dinov3 model into the ONNX format and validating its performance.

Switch the backbone of mask2former to DINOv3 for instance segmentation

[DEIMv2] Real Time Object Detection Meets DINOv3

基于 DINOv3 密集 patch 特征的子图定位工具，可快速定位子图在原图中的精确位置

Halfway to 3D: Ensembling 2.5D and 3D Models for Robust COVID-19 CT Diagnosis

[ICLR 2026] The implementation of the paper Foundation Visual Encoders Are Secretly Few-Shot Anomaly Detectors

[CVPR 2026] Ultra-Low Bitrate Perceptual Image Compression with Shallow Encoder

Winner project of the HackUDC 2026 Inditex AI challenge. From a photo of an outfit and given the Zara catalog, identify all the individual pieces the model is wearing.

[OV-DEIM] Real-time DETR-Style Open-Vocabulary Object Detection with GridSynthetic Augmentation

This is an implemetation of Dinov3 for mlx. the model is from Meta

From Misclassification to Outliers: Joint Reliability Assessment in Classification

My solution for Kaggle CSIRO Image2Biomass Challenge

Command-line tool for extracting DINOv3, CLIP, SigLIP2, RADIO, features for images and videos