Mak Gaiduk

3.04K subscribers

53 videos

Long computer vision videos with paper deep-dives and source code reads for the modern SOTA models Hey, my name is Mak, and welcome to my channel! I am deepl...

Latest Videos

Self-supervised pretraining for image models with LightlyTrain

05. Intro to Torch Lightning

04. torch.DistributedDataParallel

03. Intro to torch.distributed

02. torch.nn.DataParallel

01. Distributed training parallelism methods. Data and Model parallelism

NLP Series IV Sparse Retrieval Embeddings, Splade. Algorithm and code read.

NLP Series III Text Retrieval Embeddings, E5. Algorithm and code read.

DETR for panoptic segmentation - algorithm and code reading

Reading Segment Anything source code

RTDETR v2 - Real Time Object Detection Updates, algorithm and code reading

SAM - Segment Anything model for promptable pixel segmentation

Reading RT DETR source code

RT DETR - realtime object detection with transformers

Reading Codetr source code - part 2 out of 2

Reading Codetr source code - part 1 out of 2

CoDETR - SOTA object detection with transformers

Multiple Object Tracking Metrics - MOTA, IDF1, HOTA. Algorithm and source code reading

Reading MOTR source code

MOTR - Object Tracking with Transformers

Reading Trackformer source code

Trackformer - Multi-Object Tracking with Transformers

Reading Grounding Dino source code

Grounding Dino for open set object detection

Reading BERT source code

Reading DINO source code - DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection

Reading GPT-2 source code

DINO - DETR with Improved DeNoising AnchorBoxes for End-to-End Object Detection

Reading SWIN transformer source code - Image Recognition with Transformers

SWIN transformer (image recognition)