L = -1/N ∑ [y_i log(ŷ_i) + (1 - y_i) log(1 - ŷ_i)]Attention(Q, K, V) = softmax(QK^T / √d_k)Vf(x) = 1 / (1 + e^{-x})∇ × E = -∂B/∂tH(p, q) = -∑ p(x) log q(x)E[X] = ∫ x f(x) dxP(A|B) = P(B|A)P(A) / P(B)w^T x + b = 0θ_{t+1} = θ_t - η ∇_θ J(θ)
[HASANTAVISION]
v2.0.4 // ONLINE
{ }

MAKE YOUR APP SEE

Hover over a demo to see it live.

Back to Home
Real-time Processing // 2025-08

Object Detection & Tracking

Demo: Left Panel ←
Source Code

Multi-object tracking system designed for traffic monitoring and crowd analysis in smart city applications. The pipeline utilizes a highly optimized YOLOv8 model for detection and a custom implementation of DeepSORT for tracking across multiple camera feeds. Handles occlusions and ID switches with a custom appearance feature extractor.

Role

Computer Vision Engineer

Technologies

YOLOv8DeepSORTOpenCVC++

Use Cases & Advantages

- Smart Traffic Light Optimization - Retail Store Heatmapping and Analytics - Security Perimeter Monitoring - Smart Parking Systems Our Stack's Advantage: Optimized DeepSORT integrated directly with C++ ensures minimal ID switching across occlusions, outperforming off-the-shelf tracking libraries.