Lidar : Point Cloud
Computer vision systems aim to simulate the human visual system, allowing machines to understand or infer objects based on visual information from digital images or videos.The process involves the machine interpreting the visual information obtained by cameras, like the human brain processes visual information from the eyes to recognize objects like cats.
Understanding ...
GMSL Cameras in Robotics
both the jetson SOM and the GMSL deserializer should be powered on
Often times in robotics vision applications require high bandwidth, long-range support, data integrity, and camera reliability. MIPI CSI-2, USB 3.0, and GigE are some of the common camera interfaces in these applications. Ethernet and CAN are often used in cars but they aren’t fast enough to handle large...
Segment Anything Model — A game changer for annotation effort. A prompt based segmentation method and ways to hack it
Segment Anything:
Segment Anything segments images into masks regardless of the image type.
Previous segmentation models were application-specific (e.g., medical, pedestrian detection), but Segment Anything works universally. Earlier methods were pixel wise annotation which is time consuming. Create dataset is expensive, pixel wise annotation very expenssive. not promptable...
Semantic Segmentation: Pixel level classification
Introduction:
Image segmentation is a process that subdivides an image into it’s constituent parts of objects. It is a key task in comuter visionand image processing. It can be formulated as a pixel classification problem with 3 different approaches.
a) semantic
b) instance
c) panoptic
segmentated into colors are visualization
Examples and Applications:
a) image edi...
Edge Accelerators- solves conflict of quantization between a dynamic range and the precision
quantization techniques
Why we need quantization in edge computing?
a neuron
The edges and nodes are computational operations and there are some mappings with the hardware. But when every edge eventually has some transfer of data and that needs some storage even if temporary and very close to memories. Hence, if there is low bit, there will be less transfers amount...
Understanding why quantization is beneficial in Neural Network specially for Llama 2 or for edge computing.
Let’s start with floating point representation, it is a tradeoff between precision and long range. This requires large compute resources.
Moving from floating-point representation to quantization involves converting the continuous range of floating-point numbers to a discrete set of values represented by integers. This process is important for optimizing machine learnin...
