Results for “caption”

DreamLIP: Language-Image Pre-training with Long Captions

Scores generated image captions against reference captions and source images using the Polos metric, which is trained to align with human judgments and probes hallucination robustness and open-vocabulary evaluation.

Automate Castingwords transcription and captioning tasks through Composio's toolkit via Rube MCP.

Deploy a standalone RT-VLM dense-captioning microservice and exercise its REST API endpoints for file upload, caption generation, streaming, chat completions, and Kafka integration.

Automate Amara subtitle and captioning operations through Composio's Amara toolkit via Rube MCP.

DALL-E 3: Improving Image Generation with Better Captions

Fetches the transcript of a YouTube video as text or JSON using the video's caption track.

Write platform-specific social media captions that drive engagement and conversions. Use when the user needs compelling written content for social posts.

Evaluates image captions by converting them into scene graphs and computing an F-score over semantic propositions, measuring how well a generated caption captures the meaning of an image compared to human references.

Turn a weekly changelog markdown into a branded square video with animated UI mockups, voiceover, and captions.

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet `anchor` rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

Provides domain-specific knowledge for working with Remotion, covering 3D, animations, assets, audio, captions, and more.

Vision-language chat: VQA, captioning, image dialogue.

Exports approved Figma assets and assembles the delivery package with named files, captions, alt text, and publishing instructions.

CapsFusion: Rethinking Image-Text Data at Scale

Create and edit video compositions in React using Remotion, with guidance on layout, animation, assets, captions, effects, and rendering.

COCO: Microsoft COCO: Common Objects in Context

Generates a polished Instagram post with a hero image, caption, and hashtags based on a brief and brand style.

Analyze images using Azure AI Vision SDK for Java, enabling captioning, OCR, object detection, tagging, and smart cropping.

Packages an existing talking-head, interview, or podcast video with timed, designed graphic overlay cards—kinetic titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture—synced to the transcript, on a 16:9, 9:16, or 4:5 canvas.

Generates phrase-grounded bounding box annotations from image-caption pairs using a VLM, producing cleaned captions, referring expressions, and pixel-space bounding boxes.

Re-render a hero image into aspect ratios for Instagram, TikTok, YouTube Shorts, and Twitter/X.

LaCLIP: Improving CLIP Training with Language Rewrites

Renders a story, question, or meme as a realistic Reddit post card with vote rail, comments, and awards, suitable for video overlays or social media sharing.

Renders tweet content as a realistic X (Twitter) post card image for video overlays or image sharing, with interactive metrics and customizable themes.

Generates a Twitter share card with a hero quote, author attribution, category tag, and subtle texture, ready for screenshot and posting.

Provides detailed, structured image descriptions covering objects, people, colors, text, and scene context, with an overview and interpretation.

Generates a single-frame HTML template with a cinematic film-leak aesthetic: warm light leaks, 35mm grain, 2.39:1 letterbox, and serif typography for opening titles or chapter cards.

Scrapes posts from an Instagram user's profile feed including captions, media URLs, like/comment counts, timestamps, and location tags using cursor-based pagination.

Sigmoid Loss for Language Image Pre-Training