Pinned Repositories
6824project
final class project for MIT 6.824 (2011)
exsample
(Research) reduce the use of expensive object detectors when searching large video repos via clever sampling algorithms
frameserver
convenience tools to get frames from video for interactive visualization.
mplan2vdl
(Research) query compiler mapping logical relational query plans to optimized array operations
pgserver
Pip-installable, embedded-like postgres server for your python app
ros_access_pattern
assorted tools/useful snippets
rosbag2parquet
Rosbag2parquet transforms ROS .bag files into query friendlier .parquet files without going through python
seesaw
(Research) interactive retrieval system+algorithms: find objects of interest within image databases with less human effort
pixeltable
Data Infrastructure for Multimodal AI: Data, models, and orchestration in a unified declarative interface.
orm011's Repositories
orm011/pgserver
Pip-installable, embedded-like postgres server for your python app
orm011/seesaw
(Research) interactive retrieval system+algorithms: find objects of interest within image databases with less human effort
orm011/frameserver
convenience tools to get frames from video for interactive visualization.
orm011/exsample
(Research) reduce the use of expensive object detectors when searching large video repos via clever sampling algorithms
orm011/ros_access_pattern
assorted tools/useful snippets
orm011/plot9img
versatile visualization of images and labels (eg boxes, text, masks) by re-using grammar of graphics
orm011/basicserve
orm011/CLIP_prefix_caption
Simple image captioning model
orm011/clipexp
some clip experiments
orm011/clnexp
code for some random experiments on 'cleaning' some dataset
orm011/dsutils
some reused code
orm011/hwang
Fast sparse video decode
orm011/jsonata-wrapper
Python wrapper for the JavaScript JSONata library.
orm011/mitaskem
MIT private ASKEM repo
orm011/mmtool
some tools for dealing with text/image data
orm011/nbview
random notebooks
orm011/oldio
orm011/open_flamingo
An open-source framework for training large multimodal models.
orm011/orm011.github.io
site
orm011/pagina_ejemplo
ejemplo
orm011/pixeltable
orm011/pytorch-coviar
Compressed Video Action Recognition
orm011/skyhookml
SkyhookML is an easy-to-use web platform for computer vision.
orm011/some_nbs
orm011/supercloud_util
tools to run in sc
orm011/ta1viewer
just some viewer tool for complex output
orm011/tablespec
some experiments with modeling languages to help spec a table abstraction
orm011/testpy
test repo about python packaging
orm011/tmp
techreport temp hosting
orm011/xmodexp
random exploration cross modal