siglip 开源项目推荐（207 个）

merveenoyan /siglip

Projects based on SigLIP (Zhai et. al, 2023) and Hugging Face transformers integration 🤗

B0yko /tern

Local AI search for podcast and video archives on Apple Silicon: Whisper speech, Apple Vision OCR and SigLIP-2 visual search fused into one ranked list, with a trim editor and FCPXML export. FastAPI, vanilla JS, Tauri. Source-visible, all rights reserved.

newdee /magpie

Spotlight-style local search for everything you saved and forgot: GitHub stars, local files, images, and bookmarks. Privacy-first, no full-disk scanning, fully on-device.

qubvel /transformers-notebooks

🤗 Transformers 视觉模型的推理和微调示例

aimagelab /LLaVA-MORE

[ICCVW 2025] LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction Tuning

rizavelioglu /tryoffdiff

[CVPR'25-Demo] “TryOffDiff：通过扩散模型进行高保真服装重建的虚拟试脱”的官方仓库

MCG-NJU /AWT

[NeurIPS 2024] AWT: Transferring Vision-Language Models via Augmentation, Weighting, and Transportation

Gianluca-Improta /cinekive

Local-first cinematic still archive for filmmakers. Search your frames by look, mood, technique or director with SigLIP semantic search; ingest video and URLs; build moodboards. Free desktop app for Win/Mac/Linux, Pro unlocks studio tools. Your library stays on your disk.

ymxlzgy /FoundAD

[ICLR 2026] 论文《基础视觉编码器是秘密的少量样本异常检测器》的实现

MaviProject /MachineLearning

本项目以应用为主出发，结合了从基础的机器学习、深度学习到目标检测以及目前最新的大模型，采用目前成熟的 第三方库、开源预训练模型以及相关论文的最新技术，目的是记录学习的过程同时也进行分享以供更多人可以直接进行使用。

soloiaros /archies-appstore-lookup

Natural language Jev-powered AppStore indexing tool: describe an ambiguous feature of an app, get and research matching apps from the US top-charts. Ideal for solo developers doing market research.

miccunifi /Cross-the-Gap

[ICLR 2025] - Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversion

PRITHIVSAKTHIUR /FineTuning-SigLIP-2

Fine-Tuning SigLIP 2 for Single/Multi-Label Image Classification. Image classification vision-language encoder model fine-tuned for Image Classification Tasks

rhysdg /vision-at-a-clip

Low-latency ONNX and TensorRT based zero-shot classification and detection with contrastive language-image pre-training based prompts

awsaf49 /flickr-dataset

Download flickr8k, flickr30k image caption datasets

filipbasara0 /simple-clip

A minimal, but effective implementation of CLIP (Contrastive Language-Image Pretraining) in PyTorch

NikosEfth /freedom

Official PyTorch implementation of the WACV 2025 Oral paper "Composed Image Retrieval for Training-FREE DOMain Conversion".

hopit-ai /Moda

MODA: open fashion retrieval benchmark and models by Hopit AI. MODA (203M, open source), MODA Pro Lite (213M, open weights), MODA Pro (hosted). Full-corpus benchmarks vs FashionSigLIP, SigLIP-SO400M and ZooClaw — one harness, losses shown. #1 open model on LookBench.

Abhash-Chakraborty /Find

Self-hosted photo search: describe a picture in plain words and find it. Captions, objects, text and faces are processed on your own machine (SigLIP, YOLO, OCR, pgvector).

AaltoML /BayesVLM

Code for Post-hoc Probabilistic Vision-Language Models