works on

From the 1 of 6 linked papers with an AI index.

activity
20242026
collaborators

6 papers

cs.CV2026

Are DeepFakes Realistic Enough? Exploring Semantic Mismatch as a Novel Challenge

Sharayu Nilesh Deshmukh, Kailash A. Hambarde, Joana C. Costa +2

The paper introduces a new evaluation class for DeepFake detection that captures semantic inconsistencies between audio and video (RARV‑SMM) and tests how current models handle suc…

cs.CV2025

ZQBA: Zero Query Black-box Adversarial Attack

Joana C. Costa, Tiago Roxo, Hugo Proença +1

Current black-box adversarial attacks either require multiple queries or diffusion models to produce adversarial samples that can impair the target model performance. However, thes…

cs.CV2025

LISArD: Learning Image Similarity to Defend Against Gray-box Adversarial Attacks

Joana C. Costa, Tiago Roxo, Hugo Proença +1

State-of-the-art defense mechanisms are typically evaluated in the context of white-box attacks, which is not realistic, as it assumes the attacker can access the gradients of the…

cs.CV2024

ASDnB: Merging Face with Body Cues For Robust Active Speaker Detection

Tiago Roxo, Joana C. Costa, Pedro Inácio +1

State-of-the-art Active Speaker Detection (ASD) approaches mainly use audio and facial features as input. However, the main hypothesis in this paper is that body dynamics is also h…

cs.CV2024

BIAS: A Body-based Interpretable Active Speaker Approach

Tiago Roxo, Joana C. Costa, Pedro R. M. Inácio +1

State-of-the-art Active Speaker Detection (ASD) approaches heavily rely on audio and facial features to perform, which is not a sustainable approach in wild scenarios. Although the…

cs.CV2024

How to Squeeze An Explanation Out of Your Model

Tiago Roxo, Joana C. Costa, Pedro R. M. Inácio +1

Deep learning models are widely used nowadays for their reliability in performing various tasks. However, they do not typically provide the reasoning behind their decision, which i…