Vehicle and Pedestrian Detection Using YOLOv8 and Kitti dataset

YOLOv8, a state-of-the-art object detection algorithm, is leveraged for vehicle and pedestrian detection in real-time scenarios. Trained on the KITTI dataset, it accurately identifies vehicles and pedestrians, ensuring road safety and enhancing autonomous driving systems.
***********************************************************************
For queries: You can comment in comment section or you can email me at [e-mail]
***********************************************************************
Utilizing YOLOv8 with the KITTI dataset enables precise localization and classification of vehicles and pedestrians, even in challenging environments such as varying lighting conditions and occlusions. This robust detection system serves as a crucial component in enhancing road safety measures and advancing the development of autonomous vehicles by providing reliable perception capabilities.
#yolov8 #computervision #objectdetection #pedestrian #kittidataset

[@pifordtechnologiespvtltd5698]

Simply wonderful again. Keep Rocking

[@CodeWithAarohi]

Thank you!

[@rishisharma5827]

I love your work. Thank you so much

[@CodeWithAarohi]

Glad you enjoy it!

[@plutogaming8981]

Can you make a project , where we can find the distance of each car or pedistrian from the view of camera using open cv

[@CodeWithAarohi]

Noted!

[@CC-yy1wg]

Hello Teacher nice presentation . Am a student and also doing a small project related to tracking and detection of Multiple of objects ( Av related ) but I have been having difficulties . With some data sets and results . Is there any way I would reach you to talk more ? I will be glad on your help

[@CodeWithAarohi]

You can email me at [e-mail]

[@sanketpatil6599]

Hello Mam I am Mtech students NIT Surat.my dissertation work is in vehicle detection and tracking for Indian conditions using UAV I have done vehicle detection using Yolov8 with custom dataset of 4lak +annotations of vehicles now I am working on Tracking but couldn't able to process regarding that I want to discuss with can you help to resolve the issue

[@CodeWithAarohi]

Sure, Mail me at [e-mail]

[@sanketpatil6599]

​@@CodeWithAarohi mam I have sent you the mail

[@deepakthirukkumaran9636]

hi ma'am, can in get the colab link for the code

[@naveenkumarjadi2915]

Hi mam
can you tell me how did you converted the label files , i can filers the labels but how co-ordinates converted.

[@MahfuzurRahman-swe]

same question please help

[@g.ananthanarayanan6993]

@@MahfuzurRahman-swe did you do it? If yes how did you do it?

[@arnavthakur5409]

Nice video

[@CodeWithAarohi]

Thanks

[@automotor_in]

lots of information , thank u

[@CodeWithAarohi]

Glad it was helpful!

[@ashimasingla103]

Nice video.

[@CodeWithAarohi]

Thanks!

[@chicatechhub]

how you converted the labels from string to float like pedestrain to 0

[@Abdelkader-hq7ds]

It is done in through the yaml file

[@g.ananthanarayanan6993]

@@Abdelkader-hq7ds How do we do it? Can you explain it it detail.

[@sreedeviprasad8143]

Which is best pedestrain detection dataset?

[@CodeWithAarohi]

The "best" pedestrian detection dataset depends on your specific requirements. Some widely used datasets include COCO, Caltech Pedestrian, KITTI, INRIA Person, CityPersons, Penn-Fudan, and UA-DETRAC. Each has its own strengths and focuses, so choose based on your project's needs.

[@solomondargie5028]

Could you make video for detectron2 using kitti dataset please?

[@CodeWithAarohi]

I will try!

[@solomondargie5028]

@@CodeWithAarohi Thanks madam. and if you can do it in google colab if not it's ok using Jupiter notebook.🙏

[@NanoGi-lt5fc]

Mam plzz make end to end projects for our placement plzz with Amazon ec2 for deployment

[@CodeWithAarohi]

Sure!

[@kstwok]

Is it possible to get source code for this?

[@CodeWithAarohi]

Code is available for channel members (Contribution level 2)

[@afridirahman3906]

Can you provide the github link of this ipnyb file please.

[@CodeWithAarohi]

I don't have this code on github. You can just prepare the data.yaml file as per your dataset and execute below mentioned commands to train and test model.
Train:
from ultralytics import YOLO
# Load a model
model = YOLO('yolov8n.pt') # load a pretrained model (recommended for training)
# Train the model with 2 GPUs
results = model.train(data='coco8.yaml', epochs=100, imgsz=640)
Predict:
from ultralytics import YOLO
# Load a pretrained YOLOv8n model
model = YOLO('best.pt') # path of trained model
# Define path to the image file
source = 'path/to/image.jpg'
# Run inference on the source
results = model(source) # list of Results objects

[@skp6914]

can you share the code?