Product Finder

Product Submit

AI Models Finder

MCP Servers

MCP Client

MCP Inspector

Case Tutorials

Latest AI News

AI Daily Brief

Open-Vocabulary SAM

Interactive Segmentation and Recognition Model

CommonProductImageVision Foundation ModelInteractive Segmentation

Open-Vocabulary SAM is a vision-based foundation model built upon SAM and CLIP, focusing on interactive segmentation and recognition tasks. It achieves a unified framework of SAM and CLIP through two unique knowledge transfer modules, SAM2CLIP and CLIP2SAM. Extensive experiments on various datasets and detectors demonstrate the effectiveness of Open-Vocabulary SAM in segmentation and recognition tasks, significantly outperforming naive benchmarks combining SAM and CLIP. Moreover, training with image classification data enables this method to segment and recognize approximately 22,000 categories.

Open-Vocabulary SAM

Open-Vocabulary SAM Visit Over Time

Monthly Visits

23904807

Bounce Rate

43.33%

Page per Visit

5.8

Visit Duration

00:04:51

Open-Vocabulary SAM Visit Trend

Open-Vocabulary SAM Visit Geography

Open-Vocabulary SAM Traffic Sources

Open-Vocabulary SAM Alternatives

Open-Vocabulary SAM — Interactive Segmentation and Recognition Model

•Vision Foundation Model•Interactive Segmentation

pix2gestalt — Zero-Shot Segmentation Framework

•Image Processing•Zero-Shot Segmentation

Omni-Zero-Couples

Omni-Zero-Couples — Zero-shot stylized couple portrait creation

•Deep Learning•Image Generation

SigLIP2 — SigLIP2 is a multilingual vision-language encoder developed by Google for zero-shot image classification.

•Multilingual•Zero-shot Classification

AnimateZero — Zero-Shot Image Animation Generator

ChineseSelection

•image animation•zero-shot

SAMURAI — Zero-shot visual tracking model with motion-aware memory.

•Visual Tracking•Zero-Shot Learning

FRESCO — CVPR 2024 conference paper project, a space-time correspondence method for zero-shot video translation

•Zero-Shot Video Translation•Space-Time Correspondence

MimicBrush — Zero-shot image editing, mimic the style of reference images with one click

•Image Editing•Zero-Shot Learning

RERENDER A VIDEO — Video Rerendering: Zero-Shot Text-Guided Video-to-Video Translation

•Video•Rendering

MaskGCT — Zero-shot text-to-speech conversion model that does not require alignment information.

•Text-to-speech•Zero-shot learning

Computer Vision with DirectAI — Establish powerful computer vision models without code or training data

•Artificial Intelligence•Computer Vision

seed-vc — Zero-shot voice conversion technology that achieves high-fidelity transformation of quality and tone.

•Voice Conversion•Zero-shot Learning

X-Dyna — X-Dyna is a zero-shot human image animation generation technology based on diffusion models.

•Image Animation•Zero-Shot Learning

Diffusion Self-Distillation — A diffusion self-distillation technique for zero-shot custom image generation.

•Image Generation•Zero-shot Learning

NaturalSpeech 3 — NaturalSpeech 3 is a zero-shot speech synthesis system that utilizes a decompositional encoder-decoder and diffusion model to generate natural-sounding speech.

•Artificial Intelligence•Speech Synthesis

Whisper large-v3-turbo

Whisper large-v3-turbo — Efficient automatic speech recognition model

•Automatic speech recognition•Speech translation

GLEE — A General Object Foundation Model for Images and Videos

•Image•Video

ViTMatte — Enhanced Image Segmentation with a Pretrained Pure Vision Transformer

•Image Segmentation•Vision Transformer

VideoGrain — VideoGrain is a zero-shot method for category-level, instance-level, and part-level video editing.

•Video Editing•Diffusion Model

Segment Anything 2 for Surgical Video Segmentation — An advanced model for surgical video segmentation.

•Surgical video segmentation•Computer vision

InstantID — Zero-shot photo generation with identity preserved.

•Image Synthesis•Identity Preservation

Vision AI — Decipher valuable insights from images using AutoML Vision, leverage pre-trained Vision API models, or create computer vision applications with Vertex AI Vision

•Computer Vision•Machine Learning

seed-tts-eval — A testing dataset for evaluating a model's zero-shot speech generation capability

•Speech Synthesis•Automatic Speech Recognition

ROCKET-1 — Master the visual-temporal context prompting model for open-world interactions.

•Visual-Language Model•Embodied Decision-Making

Vision Arena — Vision Arena is an open-source platform for testing and comparing computer vision models directed to the computer vision field

•Computer Vision•Model Evaluation

Aya Vision

Aya Vision — Aya Vision is a multilingual and multimodal vision model launched by Cohere, aiming to enhance visual and text understanding capabilities in multilingual scenarios.

InternationalSelection

•Multilingual•Multimodal

whisper-diarization — Automatic speech recognition and speaker segmentation based on OpenAI Whisper

•Speech Recognition•Speaker Segmentation

ID-Animator — Zero-Shot Human Video Generation Technology

•Artificial Intelligence•Video Generation

MobileLLM-600M — An efficient and optimized 600M parameter language model designed for device applications.

•Language Model•Transformer

T-Rex2 — General vision-semantic object detection, no task-specific fine-tuning required.

•General Object Detection•Zero-Shot Detection