Niccolò Biondi

Awesome-VLA-Safety. [Arxiv] Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms

Uni-MoE. Uni-MoE: Lychee's Large Multimodal Model Family.

DepthART. [ACMMM 2026] Official code for DepthART: Depth Anything Rethought for Tiny Models

WMReward. [CVPR 2026 Highlight] Leveraging latent world model's physics understanding to improve the physics plausibility of video generation

jepa. implementing minimal versions of joint-embedding predictive architecture (JEPA)

timme. timm, evolved

FRED. FRED: The Florence RGB-Event Drone Dataset

AdvFLYP. [CVPR-26 finding 🎉] This is the official repository for our work "Finetune Like You Pretrain: Boosting Zero-shot Adversarial Robustness in Vision-language Models".

EAGC. (CVPR2026 Highlight) The Devil Is in Gradient Entanglement: Energy-Aware Gradient Coordinator for Robust Generalized Category Discovery (EAGC)

andrej-karpathy-skills. A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

claw-code. An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

SEM. [CVPR Findings 2026] SEM: Sparse Embedding Modulation for Post-Hoc Debiasing of Vision-Language Models

proactivebench. Official repository of "ProactiveBench: Benchmarking Proactiveness in Multimodal Large Language Models" (ECCV 2026)

boxmot. BoxMOT: Pluggable Python and C++ SOTA multi-object tracking modules with support for axis-aligned and oriented bounding boxes

autoresearch. AI agents running research on single-GPU nanochat training automatically

dinov2. PyTorch code and models for the DINOv2 self-supervised learning method.

lambda_orthogonality. [NeurIPS2025] λ-Orthogonality Regularization for Compatible Representation Learning

FlexOlmo. Code and training scripts for FlexOlmo

drawio-desktop. Official electron build of draw.io

open-reid. Open source person re-identification library in python

MedSAM3. MedSAM3: Delving into Segment Anything with Medical Concepts

Latent-VG. [ICCV 2025] Official code for "Latent Expression Generation for Referring Image Segmentation and Grounding"

InstanceVG. [TPAMI2025] Improving Generalized Visual Grounding with Instance-aware Joint Learning

I-JEPA. Implementation of I-JEPA from "Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture"

awesome-reid-dataset. Collection of public available person re-identification datasets

nanoGPT. The simplest, fastest repository for training/finetuning medium-sized GPTs.

nanochat. The best ChatGPT that $100 can buy.

pysparnn. Approximate Nearest Neighbor Search for Sparse Data in Python!

vllm. A high-throughput and memory-efficient inference and serving engine for LLMs

dinov3. Reference PyTorch implementation and models for DINOv3

ty-vscode. A Visual Studio Code extension for ty.

ty. An extremely fast Python type checker and language server, written in Rust.

UniReps-resources.

CrossModalityFramework. A framework to train cross modality vision models (rgb-events)

easymcp. TypeScript

CoMix. Comics Dataset Framework for Comics Understanding

LCE. Learning Compatible Embeddings, ICCV 2021

TotalSegmentator. Tool for robust segmentation of >100 important anatomical structures in CT and MR images

MONAI. AI Toolkit for Healthcare Imaging

task_adaptation. Python

Thesis. Python

raft. RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

cuvs. cuVS - a library for vector search and clustering on the GPU

Awesome-Medical-Image-Segmentation-Dataset. A list of publicly available medical image segmentation dataset.

cvat. Computer Vision Annotation Tool (CVAT) is a leading platform for building high-quality visual datasets for vision AI. It offers open-source, cloud, and enterprise products, as well as labeling services, for image, video, and 3D annotation with AI-assisted labeling, quality assurance, team collaboration, analytics, and developer APIs.

mammoth. An Extendible (General) Continual Learning Framework based on Pytorch - official codebase of Dark Experience for General Continual Learning

dS-Ease. Python

retinaface. RetinaFace: Deep Face Detection Library for Python

GhostFaceNets. This repository contains the official implementation of GhostFaceNets, State-Of-The-Art lightweight face recognition models.

ViT-Pytorch. Vision Transformer in Pytorch

cifar10-airbench. CIFAR-10 speedruns: 94% in 2.6 seconds and 96% in 27 seconds

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

compatibility-tutorial. A Tutorial on Learning Backward-Compatible Representations

CoMix-dataset. Repository for "CoMix: Comprehensive Benchmark for Multi-Task Comic Understanding"

awesome-comics-understanding. The official repo of the Comics Survey: "A missing piece in Vision and Language: A Survey on Comics Understanding"

KDPL. [ECCV 2024] - Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation

llama3-from-scratch. llama3 implementation one matrix multiplication at a time

ComicVT5. Official PyTorch implementation of Multimodal Transformer for Comics Text-Cloze

git-re-basin. Code release for "Git Re-Basin: Merging Models modulo Permutation Symmetries"

pipx. Install and Run Python Applications in Isolated Environments

bert_score. BERT score for text generation

lmdeploy. LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

LAMDA-PILOT. 🎉 PILOT: A Pre-trained Model-Based Continual Learning Toolbox

fiftyone. Refine high-quality datasets and visual AI models

iamcl2r. [CVPR 2024 Highlight] - Stationary Representations: Optimally Approximating Compatibility and Implications for Improved Model Replacements (notable top 2.8%)

Dassl.pytorch. A PyTorch toolbox for domain generalization, domain adaptation and semi-supervised learning.

NEFER. [CVPR 2023] NEFER a Dataset for Neuromorphic Event-based Facial Expression Recognition

ISSUES. [ICCVW 2023] - Mapping Memes to Words for Multimodal Hateful Meme Classification

TAPE. [WACV 2024] - Reference-based Restoration of Digitized Analog Videotapes

ARNIQA. [WACV 2024 Oral] - ARNIQA: Learning Distortion Manifold for Image Quality Assessment

QualiCLIP. Quality-Aware Image-Text Alignment for Opinion-Unaware Image Quality Assessment

UniCIL. The official code for our paper "Neural Collapse Terminus: A Unified Solution for Class Incremental Learning and Its Variants".

SLC-3DMM. Python

solo-learn. solo-learn: a library of self-supervised methods for visual representation learning powered by Pytorch Lightning

ultralytics. Ultralytics YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking

yolov5. Ultralytics YOLOv5 in PyTorch for object detection, instance segmentation, classification, training, and export.

smartaccess. A web app for user access management

TwitterRealTimeSentimentAnalysis. Lambda architecture implementation using Apache Storm, Hadoop and HBase to perform Twitter real-time sentiment analysis

Adversarial_attacks_defense. In this work the proposed defense strategy is evaluated against two black-box adversarial attacks, Hop Skip Jump and Square

Gutenberg. Gutenberg is a pipeline for training a neural network in segmenting and recognising frequent words in early printed books, in particular we focus on Gutenberg’s Bible.

Depth-of-field. JavaScript

Depth-of-field-demo. HTML

Segway. We implement a control system that stabilize the robot in its vertical position, which corresponds to the unstable equilibrium state. The hardware board used to develop the robot is a LEGO MINDSTORM. We develop also an Android app for remote control.

Keyframes-GAN. [IEEE TMM 2023] This is the official repo of the paper "Perceptual Quality Improvement in Videoconferencing using Keyframes-based GAN".

HWD. Python

ImageNetV2_pytorch. ImageNetV2 Pytorch Dataset

vit-pytorch. Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

Lens. LENS Project

big_vision. Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

Loki. A Cross-Media Search Engine

ladi-vton. [ACM MM 2023] - LaDI-VTON: Latent Diffusion Textual-Inversion Enhanced Virtual Try-On

SEARLE. [ICCV 2023] - Zero-shot Composed Image Retrieval with Textual Inversion

wandb. The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.

DataLoaders_DALI. PyTorch DataLoaders implemented with DALI for accelerating image preprocessing

RevisitingCIL. Revisiting Class-Incremental Learning with Pre-Trained Models: Generalizability and Adaptivity are All You Need (IJCV 2024)

l2p. Learning to Prompt (L2P) for Continual Learning @ CVPR22 and DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning @ ECCV22

faiss. A library for efficient similarity search and clustering of dense vectors.

open_clip. An open source implementation of CLIP.

ContinualEvaluation. [Spotlight ICLR 2023 paper] Continual evaluation for lifelong learning with neural networks, identifying the stability gap.

analog-video-restoration. [ACM MM 2022 - Demo] Restoration of Analog Videos Using Swin-UNet

CIRCO. [ICCV 2023] - Composed Image Retrieval on Common Objects in context (CIRCO) dataset

class-incremental-polytope. Python

vscode-remote-release. Visual Studio Code Remote Development: Open any folder in WSL, in a Docker container, or on a remote machine using SSH and take advantage of VS Code's full feature set.

awesome-multimodal-ml. Reading list for research topics in multimodal machine learning

joypy. Joyplots in Python with matplotlib & pandas :chart_with_upwards_trend:

hlb-CIFAR10. Train to 94% on CIFAR-10 in <6.3 seconds on a single A100. Or ~95.79% in ~110 seconds (or less!)

accelerate. 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

xformers. Hackable and optimized Transformers building blocks, supporting a composable construction.

dm_nevis. NEVIS'22: Benchmarking the next generation of never-ending learners

PaLM-rlhf-pytorch. Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

synapse. Synapse: Matrix homeserver written in Python/Twisted.

element-web. A glossy Matrix collaboration client for the web.

authentik. The authentication glue you need.

ContrastiveSupervisedDistillation. This repo contains the code of "Contrastive Supervised Distillation for Continual Representation Learning", Tommaso Barletti, Niccolò Biondi, Federico Pernici, Matteo Bruni, and Alberto Del Bimbo, ICIAP2021

ViT-pytorch. Pytorch reimplementation of the Vision Transformer (An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale)

ViT-CIFAR. PyTorch implementation for Vision Transformer[Dosovitskiy, A.(ICLR'21)] modified to obtain over 90% accuracy FROM SCRATCH on CIFAR-10 with small number of parameters (= 6.3M, originally ViT-B has 86M).

CLIP4Cir. [ACM TOMM 2023] - Composed Image Retrieval using Contrastive Learning and Task-oriented CLIP-based Features

ESA. [ECCV2022] Code for Explainable Sparse Attention for Memory-based Trajectory Predictors

google-landmark. Dataset with 5 million images depicting human-made and natural landmarks spanning 200 thousand classes.

met. A large-scale dataset for instance-level recognition for artworks is introduced.

OpenCompatible. OpenCompatible provides a standard compatible training benchmark, covering practical training scenarios.

overleaf. A web-based collaborative LaTeX editor

NeuralNetworksOnGraphs. Experiment different Neural Networks to classify graph nodes on citation networks datasets

BERT_for_ABSA. In this work (Targeted) Aspect-Based Sentiment Analysis task is converted to a sentence-pair classification task and a pre-trained BERT model is fine-tuned on it.

MANTRA-CVPR20. Official Pytorch code for MANTRA - Memory Augmented Neural Trajectory Predictor (CVPR2020)

hpdl. Jupyter Notebook

just. 🤖 Just a command runner

Good_transfer. Pretrained models for "What Makes Instance Discrimination Good for Transfer Learning?".

pytorch-metric-learning. The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.

PaddleClas. A treasure chest for visual classification and recognition powered by PaddlePaddle

Font-Awesome. The iconic SVG, font, and CSS toolkit

ProjectZeroUndub. Undub project for Tecmo's Project Zero - the EU version (the first one) for the PS2

awesome-mlss. 🤖 Machine Learning Summer School Guide

regular-polytope-networks. PyTorch code for the paper "Regular Polytope Networks" (TNNLS-2021)

pyeer. PyEER is a python package for biometric systems performance evaluation. Includes ROC, DET, FNMR, FMR and CMC curves plotting, scores distribution plotting, EER and operating points estimation. It can be also used to evaluate binary classification systems.

ffcv. FFCV: Fast Forward Computer Vision (and other ML workloads!)

sonixd. A full-featured Subsonic/Jellyfin compatible desktop music player

insightface. State-of-the-art 2D and 3D Face Analysis Project

hyperion.ng. The successor to Hyperion aka Hyperion Next Generation

podman-compose. a script to run docker-compose.yml using podman

podman. Podman: A tool for managing OCI containers and pods.

enroot. A simple yet powerful tool to turn traditional container/OS images into unprivileged sandboxes.

deepops. Tools for building GPU clusters

outline. The fastest knowledge base for growing teams. Beautiful, realtime collaborative, feature packed, and markdown compatible.

alacritty. A cross-platform, OpenGL terminal emulator.

pytorch-cifar. 95.47% on CIFAR10 with PyTorch

pytorch. Tensors and Dynamic neural networks in Python with strong GPU acceleration

DALI. A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

DeepSpeed. DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

ohmyzsh. 🙃 A delightful community-driven (with 2,500+ contributors) framework for managing your zsh configuration. Includes 300+ optional plugins (rails, git, macOS, hub, docker, homebrew, node, php, python, etc), 140+ themes to spice up your morning, and an auto-update tool that makes it easy to keep up with the latest updates from the community.

dotbot. A tool that bootstraps your dotfiles ⚡️

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

papirus-icon-theme. Pixel perfect icon theme for Linux

FaceX-Zoo. A PyTorch Toolbox for Face Recognition

deep-efficient-person-reid. Experiment about Deep Person Re-identification with EfficientNet-v2

fast-MPN-COV. @CVPR2018: Efficient unrolling iterative matrix square-root normalized ConvNets, implemented by PyTorch (and code of B-CNN,Compact bilinear pooling etc.) for training from scratch & finetuning.

deep-person-reid. Torchreid: Deep learning person re-identification in PyTorch.

incremental_learning.pytorch. A collection of incremental learning paper implementations including PODNet (ECCV20) and Ghost (CVPR-W21).

ripgrep. ripgrep recursively searches directories for a regex pattern while respecting your gitignore

podman.io_old. Repository for podman.io website using GitHub Pages.

continuum. A clean and simple data loading library for Continual Learning

aifa-scraper. Software Architectures and Methodologies exam

hydra. Hydra is a framework for elegantly configuring complex applications

face.evoLVe. 🔥🔥High-Performance Face Recognition Library on PaddlePaddle & PyTorch🔥🔥

cavaface. face recognition training project(pytorch)

openBCT. Pytorch code for Towards Backward-Compatible Representation Learning [CVPR 2020 Oral]

DeepCrawl. Deep reinforcement learning for the development of RogueLike games.

AutonomousDriving. Python