Camilo Fosco's repositories
reinforcement_learning_sepsis
Trying to come up with better policies for treating sepsis, using the MIMIC dataset and a DDQN approach with an ICNN backbone. Project built for Harvard's CS282r (Reinforcement Learning for Healthcare)
memento_keras
Repository for the Memento Project, with Keras code.
deepfake_detection
Using distortion to detect deepfakes
deepfake_caricatures_website
Website for the Deepfake Caricatures Project.
mturk-importance
Importance labeling tool
ORI-webpage
Pagina web de ORI, sistema inteligente de generacion de itinerarios.
video_generation
Vdeo generation with adversarial networks form captions and first frame. Work conducted as final project of Harvard's Data Science 2 course (CS209b)
Evolution-simulator
A virtual ecosystem where agents with a multilayer-perceptron brain compete for sustenance and evolve.
video-mem-annot
Github for the Mturk task corresponding to the first iteration of the video memorability annotation tool.
brain2video_viz_website
Visualization website for the brain2video paper
df_artifact_annot_exp
Repository for the Deepfake Artifact Annotation experiment. In this experiment, users are shown deepfakes and can annotate the parts that evidence its fakeness with a brush tool.
etm_website
Website for the ETM project
Hiat---NYUAD-Hack-2017
FinTech app for jobs and loans for refugees
LabelMeAnnotationTool
Source code for the LabelMe annotation tool.
reachannot
Repository for the reachability annotation tool. Work with @eljosephs.
semantic-segmentation-pytorch
Pytorch implementation for Semantic Segmentation/Scene Parsing on MIT ADE20K dataset
tensorflow
Computation using data flow graphs for scalable machine learning
timed_df_detection_experiment
Repository for the timed deepfake detection experiment, where participants try to detect if a video is fake when exposed to it for varying periods of time
vector-quantize-pytorch
Vector Quantization, in Pytorch
video-segment-annotation-tool
Tool to annotate video segments. Initially working towards visual grounding for goal annotation
video_sequence_tool
Simple tool that shows videos in a sequence and tracks if the user presses spacebar on any of them