Custom object detection
Models trained on your own labeled images to find the specific products or parts your workflow needs, such as rods and stems on a conveyor.
Counting items on a conveyor, vehicles at a gate or visitors at an entrance is slow and error-prone when people do it by eye. MoreYeahs object detection services build AI models that find the objects you care about in images and video, follow them as they move and count them against clear rules, giving you reliable numbers without someone watching a screen.
Hi, ask me anything about Object Detection & Counting: what it covers, how long it takes, what it costs or projects we have delivered.
AI answers can be wrong. ·
Trusted by teams in the USA, UK, Australia and India











Object detection finds what is in each image or frame and where it is. Scope is defined per project, and not every capability is needed in every solution.
Models trained on your own labeled images to find the specific products or parts your workflow needs, such as rods and stems on a conveyor.
Fast models that process each incoming frame of live video, such as CCTV streams, instead of delayed or manual monitoring.
Bounding boxes or pixel-level outlines that show exactly where each object is, for example items in a bin or on a shelf.
Multi-class models that label each detected object by type, such as separating cars, trucks, bikes and buses.
Tracking algorithms such as ByteTrack follow each object across frames with a persistent ID, so movement is understood and nothing is counted twice.
Detection, tracking and counting rules combined to count objects in an area or crossing a line.
Person detection with entry and exit logic that counts people as anonymous objects, such as visitors entering a store.
Line crossing, zones, path history and dashboards that turn recorded video into reports on movement across a site.
Learn moreObject detection finds each item, then counting rules produce a reliable total. Tracking gives each moving object one ID, so it is counted once when it crosses the line, not once per frame.
Each person is treated as an anonymous object: detected, followed briefly and counted by your rules. It produces aggregate numbers and does not identify anyone.
Which objects or people, what counts as a count, where, and what decisions the numbers support.
Camera locations, resolution, frame rate, lighting and how busy the scene gets.
Collecting representative footage and labeling the objects to detect.
Training or adapting detection models for your object categories.
Setting up localization, tracking, counting lines, zones and direction rules.
Comparing system counts with manual counts in real conditions, including peak periods.
Sending counts and events to dashboards, reports or business systems, then checking accuracy over time and retraining as conditions change.
A MoreYeahs proof of concept on conveyor video.
Rods and stems move along a conveyor and each one must be counted exactly once, even as items move and overlap.
YOLOv8 segmentation finds each rod and stem in every frame
ByteTrack gives each item a persistent ID across frames
A configurable virtual line adds one only when a tracked item crosses in the set direction
Color-coded overlays and a running total on the video
OutcomeEach item counted once as it crosses the line, with tracking IDs and the running total visible on the video.
What you can expect from working with us.
Detection, segmentation, tracking and counting logic chosen per use case, using tools our teams have applied in client work and proofs of concept.
Lines, zones and direction rules designed with your team, so every count means the same thing.
Camera placement, lighting and density are assessed before building, and counts are validated against manual counts and retrained as needed.
We start from your objects, rules and decisions. Defect checks belong to visual inspection, and safety gear to PPE detection.
Donan Consulting
“We’re most impressed by their technical expertise and problem-solving approach.”
Repeat support tickets fell by about 70%.
Flyersoft
“MoreYeahs has been a dependable partner, consistently providing strong Microsoft Dynamics 365 support.”
AbdoReview summary
The team challenges requirements instead of just following user stories, and a client portal removed 60 to 70% of support cases.
Acquiresoft SolutionsReview summary
A Dynamics CRM and SharePoint migration cut the reconciliation workload by 60%.
There is no standard price for Object Detection & Counting. These factors set the size of the work, and you get an estimate before anything starts.
More object types, or look-alike items, need more labeled data and testing
Representative footage from your real site conditions has to be gathered and labeled
Overlap, density, lighting and camera angle affect how much work accuracy takes
Lines, zones and direction rules add design and testing effort
Live processing needs faster models and suitable hardware
Sending counts and events to dashboards, reports or business systems
Checking accuracy over time and retraining as conditions change
Short answers below. For anything specific to your data or systems, ask a technical lead.
Ask a technical leadServices that build AI systems to find and label specific objects in images or video, and often to track and count them for business use.
A model trained on labeled examples scans each image or frame, draws boxes around objects it recognizes and gives each a label and confidence score.
Recognition identifies what something is. Detection also finds where each object is in the image, so it can handle many objects at once.
The system detects objects, tracks them across frames and counts them inside a zone or when they cross a line, so each object is counted once.
Yes, with suitable cameras and processing. It counts people as anonymous objects entering, leaving or occupying an area.
Tracking follows each object across frames; counting applies rules, such as line crossings, to turn those tracks into totals.
Camera angle and resolution, lighting, object size, overlap, speed, scene density and how well training data matches real conditions.
Often, if the cameras provide suitable video streams (such as RTSP) with the right angle and resolution. A site assessment confirms which cameras are usable.
Tell us what objects or people you want to detect, whether you work with images or video, whether you need counts, tracking or both, what your cameras look like and where the results should go. We will suggest a practical approach and a way to test it.
We use cookies to keep the site secure, understand how it is used, and improve it. You choose what we are allowed to load: analytics and marketing stay off until you say otherwise. Privacy Policy
Strictly necessary Always on
Required for the site to work: security checks on our forms and basic session handling. These cannot be switched off.
Analytics
Google Analytics, so we can see which pages are useful and where people get stuck. Aggregated, and never used to identify you personally.
Marketing
Lets us measure which campaigns bring people here and show you more relevant ads on other sites.
Functional
Remembers smaller conveniences, like a banner you have already dismissed, so we do not show it again.