From cde2db40e8621f14b13c6f04c3242d3dc2a12441 Mon Sep 17 00:00:00 2001 From: Xu Han Date: Thu, 24 Aug 2023 18:36:16 +0300 Subject: [PATCH] Update videos in README --- README.md | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/README.md b/README.md index 31fe2e2..8964040 100644 --- a/README.md +++ b/README.md @@ -47,7 +47,7 @@ eval "$(curl -sSL nutsh.ai/install)" # Quick Overview -https://github.com/SysCV/nutsh/assets/1113875/20d160b4-9740-4fec-b85d-be26b46f25f3 +https://github.com/SysCV/nutsh/assets/1113875/024ce405-0670-433f-a595-913fd90f769b # Diverse Annotations and Versatile Scenarios @@ -55,25 +55,25 @@ https://github.com/SysCV/nutsh/assets/1113875/20d160b4-9740-4fec-b85d-be26b46f25 In addition to standard polygon operations, Bézier curves are supported, enabling precise annotations of curved objects, such as tires. -https://github.com/SysCV/nutsh/assets/1113875/fb57016b-62b6-40d5-9637-aa2ab1e65fcc +https://github.com/SysCV/nutsh/assets/1113875/e54417d9-531a-4c09-b5f2-ab079b7cf839 ## Drawing masks Create masks with pixel-level accuracy, ensuring a perfect fit with no gaps or overlaps. -https://github.com/SysCV/nutsh/assets/1113875/b1c1e45f-493b-4a75-be55-c572774eba96 +https://github.com/SysCV/nutsh/assets/1113875/63c82054-bad1-4ccb-a671-d4ae93c829fa ## Tracking Objects Especially useful when annotating videos, the platform offers an effortless way to track objects across frames. -https://github.com/SysCV/nutsh/assets/1113875/43880c0c-94d0-4053-ac72-26128318e754 +https://github.com/SysCV/nutsh/assets/1113875/b8dba949-711d-4297-9393-dafc8d813245 ## Linking Object Parts Objects obscured by others, resulting in fragmented representation, can easily be linked to form a cohesive annotation. -https://github.com/SysCV/nutsh/assets/1113875/c42ad2f3-7f08-481c-a996-db84f6281e1a +https://github.com/SysCV/nutsh/assets/1113875/c4dc0456-6006-4ac0-abb2-abafe0c88743 # Convenient Shortcuts @@ -83,13 +83,13 @@ The platform incorporates various shortcuts, streamlining the annotation process For polygons sharing a common edge segment, one can be cloned for the other, ensuring a seamless fit without any gaps or overlaps. -https://github.com/SysCV/nutsh/assets/1113875/8ec80320-97da-45b6-9c03-db687a1f00c8 +https://github.com/SysCV/nutsh/assets/1113875/daaca148-cb02-463c-84dd-c9ff041213ff ## Interpolation Given two annotations of the same type on both a start and an end frame, heuristic interpolation can automatically generate all intermediate annotations. -https://github.com/SysCV/nutsh/assets/1113875/9b78fad0-b020-44f5-9cca-781c566bce76 +https://github.com/SysCV/nutsh/assets/1113875/c4128330-1713-4276-9ae7-aa31080d9d2f # Model-Assisted Segmentation @@ -101,20 +101,20 @@ For instance, our [SAM module](https://nutsh.ai/docs/sam) taps into the [Segment Leverage deep learning models to perform segmentation tasks across entire images, aiding the segmentation process. -https://github.com/SysCV/nutsh/assets/1113875/18b33fda-cd47-4977-a8f7-315ff8663345 +https://github.com/SysCV/nutsh/assets/1113875/e3434baa-e156-4d6d-999e-0d0a117a1b28 ## Local Smart Segmentation Direct your attention to specific regions of an image and request the model to generate segmentation predictions for that particular section. Such localized predictions often yield more detailed segmentations. -https://github.com/SysCV/nutsh/assets/1113875/d5f1b6ca-ecaa-46f3-8bfc-28666d17b338 +https://github.com/SysCV/nutsh/assets/1113875/3570752d-3b98-4a3f-adf5-66a6b50616ac # Learning from Human Feedback In addition to utilizing prompts for refining predictions, users can make subsequent adjustments to these predictions. By gathering these modifications, you can train a model that's fine-tuned to your specific needs. Our SAM module comes equipped with features that assist in [fine-tuning the SAM decoder](https://nutsh.ai/docs/SAM%20Module#sam-decoder-fine-tuning) seamlessly. -https://github.com/SysCV/nutsh/assets/1113875/2b742260-f10d-4655-8b86-06e68a36fbe5 +https://github.com/SysCV/nutsh/assets/1113875/79d8c583-a1d3-4097-b6b4-810df7e6d7a1 # Next Steps