GitHub

RadarMOTR: Multi-Object Tracking with Transformers on Range-Doppler Maps

RadarMOTR is a transformer neural network based on MOTRv2 able to track multiple objects in Range-Doppler maps. Objects are first detected by an external detector based on CFAR & DBSCAN clustering.
GitHub

Bytemap contains types for making maps from bytes to bool, integer, or float using a backing array.

Micro-benchmarks are usually not a good way to evaluate systems. That said, using a bytemap array can be very fast while also providing readable code. Let's say you want to test that string contains ...
IEEE

Visualization of Salient Object With Saliency Maps Using Residual Neural Networks

Abstract: Visual saliency techniques based on Convolutional Neural Networks (CNNs) exhibit an excessive performance for saliency fixation in a scene, but it is harder to train a network in view of ...
IEEE

osmAG-LLM: Zero-Shot Open-Vocabulary Object Navigation via Semantic Maps and Large Language Models Reasoning

Abstract: Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features, achieving a high level of detail and guiding robots to find objects ...