Repository Issue Activity (beta)

#2353 Issue with importing other trackers

#2334 ModuleNotFoundError: No module named 'boxmot.data'

#1707 Add Selective Mechanism for Feature Extraction

#2331 Version is wrong

#2332 Instantiation of ReIDModelmodule results in ModuleNotFoundError

#2313 MOT20-ablation问题

#2290 Support for NVIDIA DeepStream on Jetson Edge Devices

#2303 why use MMOT

#2299 BoT-SORT suddenly changes multiple player IDs between consecutive frames without camera movement

#1860 done

#2287 Boxmot version mismatch

#2263 numpy.linalg.LinAlgError: 1-th leading minor of the array is not positive definite

#2274 Discussion needed

#2260 Feature Request: Support fine-tuning for RF-DETR with ByteTrack

#1988 logger.remove() in __init__.py impacts downstream projects

#2258 No results summary

#2245 Support TensorRT Engine Format for RF-DETR in BoxMOT

#2244 Why is the MOT tracking evaluation running only on few frames of sequence?

#2246 How to save the labels and the confidence score etc. during the boxmot track?

#2240 Python 3.8 Compatibility

#2207 [Bug]: TypeError: only 0-dimensional arrays can be converted to Python scalars (NumPy 2.x compatibility)

#2233 Class names are not the same as the class which the detector is trained on

#2231 --dynamic export fails on CUDA (device 0); CPU export works

#2230 ReID export with --half fails on CUDA: dummy input FP32 but model weights FP16 during warmup (FloatTensor vs HalfTensor)

#2232 Make sure engine conversion works fine