PRIS-CV: Computer Vision Group
Pinned Loading
Repositories
- FlexiGrad Public
We propose FlexiGrad, a simple and parameter-free method for hierarchical fine-grained classification that regulates gradient interactions during backpropagation.
- DataEvolver Public
Autonomous synthetic data construction via VLM-guided iterative rendering — let your data build and improve itself.
- ColourCrafter Public
[ECCV 2026] Code release for "Recolour What Matters: Region-Aware Colour Editing via Token-Level Diffusion"
- Hepto-LLaVA Public
[MICCAI 2026] The official code for "Hepato-LLaVA: An Expert MLLM with Sparse Topo-Pack Attention for Hepatocellular Pathology Analysis on Whole Slide Images"
- SSDiff Public
[NeurIPS 2025] Self-Supervised Selective-Guided Diffusion Model for Old-Photo Face Restoration
- MedReasoner Public
[AAAI 2026] The official code for "MedReasoner: Reinforcement Learning Drives Reasoning Grounding from Clinical Thought to Pixel-Level Precision"
- DDG Public
Official PyTorch implementation of **"Implicit Diffusion Distillation with Domain Disentanglement for Domain Generalization"** (ICME 2026).
- Semantic_Memory Public
- MMBench-Live Public
- VGIF-SCORE Public
[PRCV 2026] Fine-grained benchmark and evaluation framework for text-to-video generation using ST-DAG prompt decomposition, VLM-based QA, and objective/subjective scoring.
Top languages
Loading…
Most used topics
Loading…