Tag
8 English Kalera News articles tagged Computer Vision — source-backed.
Google DeepMind demonstrates the contextual understanding of an AI-powered mouse cursor, turning a handwritten note into a to-do list or booking a restaurant table directly from a video.
A developer has successfully ported the SAM 2.1 (Segment Anything Model) to Apple's MLX framework, boosting performance by 25% on Mac computers.
Pinterest has achieved a major breakthrough in operational efficiency, slashing AI infrastructure costs by 90% and boosting accuracy by 30% by restructuring the vision processing layer of the Qwen3-VL model.
NVIDIA has achieved a breakthrough in computer vision, speeding up bounding box detection by 10x by eliminating a traditionally mandatory step in VLM grounding.
NVIDIA's research team has announced LocateAnything, a new vision-language model that redefines bounding box prediction. This is a major breakthrough that enables AI agents and robots to not only 'see' but also localize objects at lightning speed for precise action.
Roboflow's RF-DETR model is now integrated into the Transformers library, enabling real-time object detection with accuracy that outperforms YOLO.
Apple introduces TC-JEPA, a new self-supervised method that uses text captions to guide and reduce noise during AI image recognition learning.
Ethan Lauer, a software engineer at Boston Dynamics, shares details on how the Stretch robot perceives the world, builds environmental models, and overcomes sensing challenges.
