Kevin Zakka

Elite
@kevinzakka

PhD @ UC Berkeley.

mink. Python inverse kinematics based on MuJoCo.

1.4k

spatial-transformer-network. A Tensorflow implementation of Spatial Transformer Networks.

1k

pytorch-goodies. PyTorch Boilerplate For Research

607

recurrent-visual-attention. A PyTorch Implementation of "Recurrent Models of Visual Attention"

476

mjctrl. Minimal, clean, single-file implementations of common robotics controllers in MuJoCo.

439

torchnca. A PyTorch implementation of Neighbourhood Components Analysis.

398

obj2mjcf. A CLI for processing composite Wavefront OBJ files for use in MuJoCo.

367

mujoco_scanned_objects. MuJoCo Models for Google's Scanned Objects Dataset

283

clip_playground. An ever-growing playground of notebooks showcasing CLIP's impressive zero-shot capabilities

178

torchkit. Research boilerplate for PyTorch.

150

tsne-viz. Python Wrapper for t-SNE Visualization

136

ibc. A PyTorch implementation of Implicit Behavioral Cloning

113

form2fit. [ICRA 2020] Train generalizable policies for kit assembly with self-supervised dense correspondence learning.

84

blog-code. My blog's code repository.

76

learn-linalg. Learning some numerical linear algebra.

72

mjc_viewer. A browser-based 3D viewer for MuJoCo

63

robopianist-rl. RL code for training piano-playing policies for RoboPianist.

60

x-magical. [CoRL 2021] A robotics benchmark for cross-embodiment imitation.

60

torchsdf-fusion. Benchmarking PyTorch variants of TSDF fusion.

50

mujoco_cube. A 3x3x3 puzzle cube model for MuJoCo.

36

robopianist-demo. C

26

learn-ransac. Learning about RANSAC.

21

dm_env_wrappers. Standalone library of frequently-used wrappers for dm_env environments.

19

nanorl. A tiny reinforcement learning codebase for continuous control, built on top of JAX.

15

software. My open-source software contributions.

9

awesome-robot-descriptions. A curated list of awesome robot descriptions

7

mujoco_utils. Python

6

mujoco_menagerie. A collection of high-quality models for the MuJoCo physics engine, curated by DeepMind.

4

learn-volumetric-fusion. Learning about volumetric fusion.

3

learn-blur. Learning about various image blurring techniques.

3

robot_descriptions.py. Import robot descriptions as Python modules

3

rscope. RL training visualizer for Mujoco Playground environments + Brax training

2

whole_body_tracking. Python

2

brax. Massively parallel rigidbody physics simulation on accelerator hardware.

1

mujoco. Multi-Joint dynamics with Contact. A general purpose physics simulator.

1

controls. Toy implementations of algorithms for path planning, system modelling, optimal control, estimation, and more.

1

mujoco_warp. GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.

1
37
Apply