T R A N S F O R M I N G

Mirco Planamente
VISION
Into
INTELLIGENCE

[online] · open to collaboration

Mirco Planamente

AI Researcher  ×  Computer Vision  ×  PhD


/// TECHNICAL STACK

Core competencies spanning deep learning architectures, computer vision pipelines, and production-grade ML frameworks.

[CORE::01]

Deep Learning

[VISION::02]

Computer Vision

[SPECIALTY::03]

Action Recognition

[ADAPT::04]

Domain Adaptation

[FUSION::05]

Multi-Modal AI

[FRAMEWORK::06]

PyTorch Expert


$ research --areas

What I Do

[01] CORE_RESEARCH
🎥

Egocentric Vision

First-person action recognition & understanding from wearable cameras

CVPR • WACV • IEEE_RA-L

[02] ADAPTATION
🤖

Domain Adaptation

Building robust AI systems that generalize across environments

Cross-domain • Transfer_Learning

[03] MULTI-MODAL
🎦

Multi-Modal Learning

Integrating audio-visual signals for comprehensive scene understanding

Audio+Vision • Sensor_Fusion

[04] APPLICATIONS
🏭

Industrial Vision

Semantic segmentation, object detection & anomaly detection for production

Real-world • Production_Systems


RESEARCH_IMPACT

26+ publications in top-tier venues including CVPR, IEEE RA-L, and WACV

PhD @ Politecnico di Torino • Computer Science & AI

$ ls ./speaking_events

Featured Talks

[EVENT_01] 2024

Py4AI Conference 🐍

Presented cutting-edge research on egocentric vision and deep learning applications to the Python AI community.

Py4AI Conference poster
Codemotion 2025 talk
[EVENT_02] 2025

Codemotion Conference 🎤

"From pixels to features: one backbone to rule them all!" A journey from revolutionary features to foundation models, exploring how backbones transform pixels into reusable feature spaces.

[ACHIEVEMENT] 2021–2022

EPIC-KITCHENS Challenge 🏆

Two-time consecutive winner of the Unsupervised Domain Adaptation Challenge. 3rd place in 2021 and Top 3 in 2022 at CVPR, competing against leading research teams worldwide.

🏆 Top 3 Winner 2021 & 2022 EPIC-KITCHENS-100

$ metrics --output

Impact by Numbers

mirco@research:~$ echo "Interested in collaboration?"
→ Let's connect and build something innovative together.

$ grep --selected publications.bib

Selected Publications

    $ connect --social