Hugging Face's logo
Vishakha Gupta's picture
2

Vishakha Gupta

vishakha041
·

AI & ML interests

Multimodal AI, graph databases, vector databases

Recent Activity

reacted to dmoxy's post with 🚀 5 days ago
☀️ The Summer of Workflows continues at ApertureDB! New release: Extract Text From Images & PDFs→ store → embed → search. 🎬 See It In Action : https://youtu.be/Q7xHoh6JIxM Features: Works with Tesseract & EasyOCR Extracts text segments + embeddings Links text back to images/PDFs in the DB Enables semantic search across multimodal data Sample use cases: Search scanned PDFs & archives Detect text in signs, labels, or documents Build retrieval systems for compliance, research, or enterprise AI apps Try it now → https://shorturl.at/Bc0eS Read The Docs: https://shorturl.at/nMUMD Get The Code: https://shorturl.at/xKgkX Additional Information: https://shorturl.at/b8InC
reacted to dmoxy's post with 🚀 5 days ago
🌞 Our Summer of Workflows finale! New release Embedding Generation for Videos. With our latest ApertureDB AI workflow, you can now generate embeddings for video frames & clips and store them directly in a multimodal database—ready for semantic search, RAG, or agentic use cases. 🎬 See It In Action: https://youtu.be/X2ZXE0EEAkk 🔎 Example use cases: Natural language search across video libraries Highlight reel creation & scene retrieval Safer content moderation Lecture indexing + video-to-text alignment Multimodal RAG (text + images + video) ✨ Check out the demo notebook to see how you can: Import videos from S3 Generate embeddings per frame/clip Query videos with natural language (“show me a baby”) 👉 Try it here: https://shorturl.at/jSNtu Turn your video library into a fully searchable knowledge base. No scrubbing—just instant, semantic video results. We’d love your feedback and ideas on how you would use it.
reacted to dmoxy's post with 😎 22 days ago
☀️ The Summer of Workflows is in the home stretch — and we are down to the final 3 releases! This week we spotlight Face Detection. 🎬 See It In Action: https://youtu.be/99T19oTq37E From photo apps to security systems, face detection is everywhere. But how do you actually build a scalable pipeline that can juggle text, images, audio, video, and more without the headaches and big bills? Introducing ApertureDB Face Detection Workflow. With this workflow you can: ✅ Detect and retrieve faces ✅ Easily add bounding boxes to faces detected in ApertureDB ✅ Extract descriptors ✅ Run similarity searches ✅ All inside a multimodal database No complex pipelines. No heavy lifting. Just plug in the workflow and go. 👉 Try It Now! https://cloud.aperturedata.io/login Read The Docs: https://shorturl.at/6ZogI Explore The Code: https://shorturl.at/W78E1 Get Additional Resources: https://shorturl.at/9T1OR 👇 Tell us what you think about the series in the comments below! We are listening.
View all activity

Organizations

ApertureData's profile picture