Skeleton-based Action Recognition

Recognizing the actions performed by humans from their skeletons.

Detecting 32 Pedestrian Attributes for Autonomous Vehicles

Joint pedestrian detection and attribute recognition with fields and Multi-Task Learning.


Human Pose Estimation

PifPaf: Composite Fields for Human Pose estimation, CVPR’19

Object detection

Adapting fields for detection from aerial images

3D human detection

We have been exploring how to detect humans in the 3D space only using cameras, which are cheap, reliable and ubiquitous. Our major applications are autonomous vehicles and delivery robots. We focused on challenging cases (the long tail) and uncertainty estimation to improve the reliability of autonomous systems.


Multi-Task Learning with Auxiliary Tasks

Revisiting Multi-Task Learning with ROCK: a Deep Residual Auxiliary Block for Visual Detection, NeurIPS’18

Visual Re-Identification

Deep Visual Re-Identification with Confidence, 2020

Super resolution & Style transfer

Perceptual Losses for Real-time Style Transfer and Single Image Super-Resolution, ECCV’16



Open Source Unified Library for Vehicle Trajectory prediction (multiple datasets, multiple SOTa models, …)



Open Source Library for Human Trajectory prediction (e.g., official code for Social LSTM, Social Gan, etc…)

S-ATTack: Analyzing Trajectory Prediction Models

We introduce a socially-attended attack to assess the social understanding of prediction models.

Social GAN

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks, CVPR’18



A vehicle trajectory prediction model which leverages both knowledge and data.

Pedestrian Stop and Go

Predicting whether pedestrians will stop walking (Stop) or start to walk (Go) in the near future, for better trajectory prediction around road traffic.

Pedestrian Bounding Box Prediction

A libary for predicting 2D and 3D bounding boxes of humans in autonomous driving scenarios


Crowd-Robot Interaction

Crowd-Robot Interaction: Crowd-aware Robot Navigation with Attention-based Deep Reinforcement Learning, ICRA’19

Generative models

Semantically-aware Discriminators

We build on successful cGAN models to propose a new semantically-aware discriminator that better guides the generator. We aim to learn a shared latent representation that encodes enough information to jointly do semantic segmentation, content reconstruction, along with a coarse-to-fine grained adversarial reasoning.

Collaborative Sampling in Generative Adversarial Networks

We propose a collaborative sampling scheme between the generator and discriminator for improved data generation. Guided by the discriminator, our approach refines generated samples through gradient-based optimization in the data (or feature / latent) space, shifting the generator distribution closer to the real data distribution.

Generating SVG

DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation

Discrete Choice Models and Neural Networks

Discrete Choice Models and Neural Networks

Code for our new choice model referred to as the Learning Multinomial Logit (L-MNL)


Ultimate labeling tool for videos

A multi-purpose Video Labeling GUI in Python with integrated SOTA detector and tracker. Developed using PyQt5