Computer Vision and Analytics

Minnesota, USA
Joined September 2022
RidgeRun.ai retweeted
RF-DETR keypoints is finally out preview release: real-time transformer keypoint detection Apache 2.0 71.8 AP on COCO, 9.7ms on T4. outperforms YOLO11-pose and YOLO26-pose at similar latency
31
152
12
1,318
98,989
RidgeRun.ai retweeted
🎉🔔RidgeRun is showcasing the Holoscan Sensor Bridge from #NVIDIARobotics and #Lattice Semiconductor at the @NVIDIAGTC Conference!🚀 🚀 🎉This technology enables low-latency video capture for real-time applications in Healthcare, Automotive, and beyond.🚀 🔍⏩Looking to push the boundaries of real-time imaging and #AI? Visit us at Booth #2212 to see this innovation in action and discuss potential applications with our experts.👩🏻‍🏫 🌐🔗RidgeRun Contact : lnkd.in/edbdvfnW📈 #EdgeComputing #LowLatency #NVIDIA #GTC2025 #LatticeSemiconductor #FPGA #GTC #Jetson #JetsonOrin @latticesemi #RidgeRun #RidgeRunAI
2
1
1
738
RidgeRun.ai retweeted
🚀At NVIDIA GTC 2025, @RidgeRunAI will showcase an AI assistant with vision and speech capabilities running on NVIDIA Jetson Orin™! This system can guide, correct, and train users in real-time, making it suitable for education, manufacturing, consumer electronics, and robotics applications.🚀 🎯Live Demo at GTC 2025: The AI helps a human player assemble a Connectix figure, handling real-world challenges like mistakes, missing parts, and problem-solving—just like in actual work environments! 🚀 ✅This technology can enhance training, automation, and error correction across industries. Imagine AI-powered hands-free guidance on assembly lines, robotics training, and more!🚀 ⏩Register to NVIDIA GTC 2025 to see it in action: lnkd.in/eUsYqEZ2 ⏩Let's meet and schedule a meeting with our team: lnkd.in/edbdvfnW #GTC2025 #AI #ComputerVision #NVIDIAJetson #EdgeAI @NVIDIAGTC @nvidia #jetson #robotics #NVIDIARobotics @NVIDIAAI
2
2
1
158
I'm at @Ai4Conferences! Hit me up if you are around.
1
61
RidgeRun.ai retweeted
Working with cutting edge technology has never been more exciting! As @nvidia partners we get to work first hand with amazing hardware and deliver the very best to our customers 🚀 #iot #ai #artificalintelligence #jetson
4
14
631
This is scary good! And @OnusFM made it open source: github.com/suno-ai/bark – at San José, Costa Rica
“Bark” is a new open source text-to-speech model that emphasizes more natural sounding imperfections in human speech Listen to the demo:
1
2
2
254
The team at @StabilityAI just open sourced their LLM: StableLM. Check out their announcement: stability.ai/blog/stability-… and the code at GitHub: github.com/Stability-AI/Stab… – at San José, Costa Rica
1
1
1
102
Thanks to the awesome @DVCorg community for the swag! You’re truly a #MLOps gem.
3
2
13
3,407
Claims to have lower word-error-rate than OpenAI’s Whisper – at San José, Costa Rica
MuAViC is a new benchmark for audio-visual learning for robust speech translation. Using MuAViC, we trained our AV-HuBERT model to translate speech in challenging settings using sound + sight, achieving results outperforming other leading translation models. More details ⬇️

1
157
RidgeRun.ai retweeted
Best of AI Twitter (Jan 25-31): - Hungry, Hungry, Hippos - a transformer killer? - 3 methods to detect and watermark LLMs, - Carper's open-source diff codegen models, - Text-to-4D & MusicLM, - Hugging Face announces a Kaggle competitor, ... and more: 1/14
4
51
1
394
134,449
Google Research using a Language Model to generate music from text. While not a new idea, definitely a huge leap.
MusicLM: Generating Music From Text abs: arxiv.org/abs/2301.11325 project page: google-research.github.io/se…
1
1
193
RidgeRun.ai retweeted
Best of AI Twitter - January 11-25 (megathread): Andrej Karpathy builds "GPT from Scratch", A class-action lawsuit against Stability AI, Track human pose using WiFi alone, An LLM-native IDE, InstructPix2Pix & AdA, Training a world-class Rocket-league RL bot, ... and more: 1/22
18
235
15
1,708
500,370
Check out this impressive zero-shot speech synthesizer that @Microsoft just announced. Looks like they are concerned about the potential implications so they haven’t released the model yet. VALL-E: valle-demo.github.io
1
2
137
⁦@DVCorg⁩ you just made our PRs 100x more valuable 🙌
RidgeRun.ai retweeted
AudioGen: Textually Guided Audio Generation Presents an autoregressive generative model to generate audio samples from text inputs. Impressive!🤯 It's incredible to see the fast acceleration of applications around generative AI. felixkreuk.github.io/text2au…
3
27
1
127