MiPa — Mixed Patch Object Detection
Computer Vision ResearcherImpact
- Critiqued and reproduced WACV 2025 paper — identified that Uniform(0,1) mixing ratio is suboptimal for modality-agnostic detection
- Proved Beta(2,2) distribution outperforms paper default by +1.24% RGB AP50 on CNN backbones with 6-epoch training
- Implemented MiPa patch mixing on both CNN (ResNet50-FPN) and ViT (Swin-Tiny) backbones using vectorized tensor operations
- Achieved 86.40% Avg AP50 on LLVIP pedestrian detection with only 4K training samples and 12.8-minute training time
Skills
Project Abstract & Details
A research critique and experimental reproduction of the WACV 2025 paper 'Mixed Patch Visible-Infrared Modality Agnostic Object Detection.' Implemented and compared 5 mixing-ratio distributions (Uniform, Beta(2,2), Beta(0.5,0.5), Truncated Uniform, Gaussian) across CNN (ResNet50-FPN) and ViT (Swin-Tiny) backbones for pedestrian detection on the LLVIP dataset. Key finding: Beta(2,2) consistently outperforms the paper's default Uniform(0,1) distribution — achieving +1.24% RGB AP50 improvement on CNN backbones. Note: Due to Kaggle GPU session constraints, CNN backbone results were partially saved; the ViT backbone experiments have complete result logs with consistent improvement across all mixing distributions.
Key Features
5 Distribution Comparison
Uniform, Beta(2,2), Beta(0.5,0.5), Truncated Uniform, and Gaussian mixing ratios compared systematically
Dual Backbone Validation
CNN (ResNet50-FPN with Faster R-CNN) and ViT (Swin-Tiny) — cross-architecture validation of all mixing strategies
WACV 2025 Reproduction
Independent critique and experimental reproduction finding Uniform(0,1) suboptimal vs Beta(2,2)
Beta(2,2) Proven Best
+1.24% RGB AP50 improvement over paper default on CNN; consistent gains across all ViT distributions tested
LLVIP Dataset
12,025 paired visible-thermal (RGB-IR) pedestrian images — 4K training samples with zero-shot evaluation
Kaggle GPU Pipeline
Vectorized tensor operations for efficient patch mixing — 12.8-minute training on single GPU (T4 x2)
Architecture
MiPa (Mixed Patch) mixing layer inserted at the input stage: visible (RGB) and infrared (IR) images are divided into patches, then mixed using a sampling distribution to create modality-agnostic training samples. Distribution parameter p controls the ratio of visible-to-infrared patches. The mixing layer is agnostic to the downstream detector — tested with both Faster R-CNN (ResNet50-FPN) and ViT-based (Swin-Tiny) detection heads. Experiments run on Kaggle GPUs (Tesla T4 x 2) with mixed-precision training over 6 epochs.
Architecture diagram placeholder