Picture for Lorenzo Seidenari

Lorenzo Seidenari

Depth-based Privileged Information for Boosting 3D Human Pose Estimation on RGB

Add code
Sep 17, 2024
Viaarxiv icon

Are CLIP features all you need for Universal Synthetic Image Origin Attribution?

Add code
Aug 17, 2024
Viaarxiv icon

Deepfake detection by exploiting surface anomalies: the SurFake approach

Add code
Oct 31, 2023
Viaarxiv icon

FLODCAST: Flow and Depth Forecasting via Multimodal Recurrent Architectures

Add code
Oct 31, 2023
Viaarxiv icon

Addressing Limitations of State-Aware Imitation Learning for Autonomous Driving

Add code
Oct 31, 2023
Viaarxiv icon

DiffDefense: Defending against Adversarial Attacks via Diffusion Models

Add code
Sep 07, 2023
Viaarxiv icon

Robot Pose Nowcasting: Forecast the Future to Improve the Present

Add code
Aug 24, 2023
Viaarxiv icon

Forecasting Future Instance Segmentation with Learned Optical Flow and Warping

Add code
Nov 15, 2022
Viaarxiv icon

Online Deep Clustering with Video Track Consistency

Add code
Jun 07, 2022
Figure 1 for Online Deep Clustering with Video Track Consistency
Figure 2 for Online Deep Clustering with Video Track Consistency
Figure 3 for Online Deep Clustering with Video Track Consistency
Figure 4 for Online Deep Clustering with Video Track Consistency
Viaarxiv icon

SMEMO: Social Memory for Trajectory Forecasting

Add code
Mar 23, 2022
Figure 1 for SMEMO: Social Memory for Trajectory Forecasting
Figure 2 for SMEMO: Social Memory for Trajectory Forecasting
Figure 3 for SMEMO: Social Memory for Trajectory Forecasting
Figure 4 for SMEMO: Social Memory for Trajectory Forecasting
Viaarxiv icon