Stars
nvidia-modelopt is a unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for do…
[ICML2025] VARSR: Visual Autogressive Modeling for Image Super Resolution
PPL Quantization Tool (PPQ) is a powerful offline neural network quantization tool.
[ACM MM 2024] QPT V2: An MIM-based pretraining framework for IQA, VQA, and IAA.
Official Code for Stable Cascade
[NeurIPS 2024 Best Paper][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ult…
https://wavespeed.ai/ Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.
The official repo of Qwen-VL (通义千问-VL) chat & pretrained large vision language model proposed by Alibaba Cloud.
①[ICLR2024 Spotlight] (GPT-4V/Gemini-Pro/Qwen-VL-Plus+16 OS MLLMs) A benchmark for multi-modality LLMs (MLLMs) on low-level vision and visual quality assessment.
The first challenge on short-form video quality assessment
Code for paper 'QNCD: Quantization Noise Correction for Diffusion Models'
[ECCV2024] XPSR: Cross-modal Priors for Diffusion-based Image Super-Resolution
SUPIR aims at developing Practical Algorithms for Photo-Realistic Image Restoration In the Wild. Our new online demo is also released at suppixel.ai.
Generative Models by Stability AI
[CVPR2024] SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
Code and documents of LongLoRA and LongAlpaca (ICLR 2024 Oral)
Real-time face swap for PC streaming or video calls
[CVPRW 2023] Zoom-VQA: Patches, Frames and Clips Integration for Video Quality Assessment
We extend Segment Anything to 3D perception by combining it with VoxelNeXt.
VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking (CVPR 2023)
[ECCV2022, TPAMI2023] FAST-VQA, and its extended version FasterVQA.
Unofficial PyTorch implementation of Masked Autoencoders Are Scalable Vision Learners
Perceptual video quality assessment based on multi-method fusion.
ncnn is a high-performance neural network inference framework optimized for the mobile platform
[official] Unified Quality Assessment of In-the-Wild Videos with Mixed Datasets Training (IJCV 2021)