๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (1ํธ)](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Acaf6645f-6ec9-4107-b539-bdc8ed1aea55%253Aimage.png%3Ftable%3Dblock%26id%3D399623bd-cb86-80de-af24-d6e2f1509d16%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (1ํธ)
2026๋
7์ 7์ผ
๐ป Profile
Jungminii
Student studying A.I, especially in Conputer Vision.
The original tech blog is on velog. This space includes daily posts.
๐ Search
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (2ํธ)](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ad8497c53-60ba-48c2-a60b-540df1e19bfb%253Aimage.png%3Ftable%3Dblock%26id%3D399623bd-cb86-8031-a70d-e86d0929ae8d%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (2ํธ)
2026๋
7์ 7์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (3ํธ)](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ab585c086-8bd7-49ef-8205-696b186f03bd%253Aimage.png%3Ftable%3Dblock%26id%3D399623bd-cb86-80b4-bb33-d625e819e047%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Open-Det: An Efficient Learning Framework for Open-Ended Detection (3ํธ)
2026๋
7์ 7์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] GenerateU: Generative Region-Language Pretraining for Open-Ended Object Detection](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A178ad4d6-1ad4-44a2-96ee-bce4ce39347e%253Aimage.png%3Ftable%3Dblock%26id%3D398623bd-cb86-80e4-96fe-f26adeaccac5%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] GenerateU: Generative Region-Language Pretraining for Open-Ended Object Detection
2026๋
7์ 6์ผ
๐ Daily
[์ธํด ์ผ์ง] ETRI: ์ธํด OT ๋ฐ Seher์์ ๋ง๋จ!
2026๋
7์ 4์ผ
ETRI ์ ์ ์ธํด ์์ ์ , OT ๋ฐ ๋ฐ๋์์ด ์์์ต๋๋ค. ๊ฐ๋จํ ๋ฐ๋์ ์ดํ, ์ฌ์์ธ Seher ์ฐ๊ตฌ์๋๊ณผ์ ์งง์ ๋ง๋จ!
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] VL-SAM: Training-Free Open-Ended Object Detection and Segmentation via Attention as Prompts](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A5feb5611-0bfb-4fc7-aef7-525f8df78235%253Aimage.png%3Ftable%3Dblock%26id%3D397623bd-cb86-808b-a86d-e40879b4dc6c%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] VL-SAM: Training-Free Open-Ended Object Detection and Segmentation via Attention as Prompts
2026๋
7์ 4์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] CLIP: Learning Transferable Visual Models From Natural Language Supervision](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Af3176ede-cdb8-49d0-b740-85ce46b94045%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80e9-8dc9-d1689926fb9a%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] CLIP: Learning Transferable Visual Models From Natural Language Supervision
2026๋
7์ 2์ผ
CV์ ํจ๋ฌ๋ค์์ '๊ณ ์ ๋ ๋ ์ด๋ธ ์งํฉ'์ ๋ํ ์ง๋ ํ์ต์์, ์น ์ค์ผ์ผ์ (์ด๋ฏธ์ง, ํ ์คํธ) ์์ ํ์ฉํ '์์ฐ์ด ๊ฐ๋ '์ผ๋ก ์ ํ์ํจ ์ ๊ตฌ์ ์ธ ์ฐ๊ตฌ์ด๋ค. ์ ๋ก์ท(Zero-shot) ์ ์ด ํ์ต์ ์๋ก์ด ๊ฐ๋ฅ์ฑ์ ์ด์์ผ๋ฉฐ, ํ๋กฌํํธ ์์ง๋์ด๋ง์ ํฌํจํ ๋ฐฉ๋ฒ๋ก ์ ์ ์ํ๋ค
๐ Paper Review
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Beyond-Labels: Advancing Open-Vocabulary Segmentation With Vision-Language Models
2026๋
7์ 1์ผ
๐ Paper Review
[ ๋ ผ๋ฌธ ๋ฆฌ๋ทฐ ] CrackSSM: Reviving SSMs for Crack Segmentation via Dynamic Scanning
2026๋
6์ 18์ผ
๐ Study
Mask ์ด๋ฏธ์ง ์ฒ๋ฆฌ์ ์ฃผ์ํ ์
2026๋
6์ 16์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋๊ตฌํ] Symmetry-Aware Transformer-based Mirror Detection](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A8b90141a-b1d0-44f0-a75f-c2e089289e54%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80ab-86fc-fb2c570eb17a%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋๊ตฌํ] Symmetry-Aware Transformer-based Mirror Detection
2026๋
6์ 14์ผ
๐ Study
Dot product, Inner product ๊ทธ๋ฆฌ๊ณ element-wise
2026๋
6์ 7์ผ
๐ Study
optimizer์ ์ํ์ ์ถ๋ก
2026๋
5์ 23์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ]: Agent Laboratory](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A61af60ac-a8aa-448f-8714-51dafc64a89c%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-800a-ab14-e7d9d1613153%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ]: Agent Laboratory
2026๋
5์ 13์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] ConvNeXt](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A07b262b9-aa80-4b65-a93a-0962f535e117%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-802f-a2ba-c763b47efca6%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] ConvNeXt
2026๋
5์ 4์ผ
๐ Study
[ ๋ ผ๋ฌธ ๋ฆฌ๋ทฐ ] The Facial Action Coding System: Principles and Applications
2026๋
3์ 29์ผ
๐ Paper Review
[ ๋ ผ๋ฌธ ๋ฆฌ๋ทฐ ] Recurrent Convolutional Strategies for Face Manipulation Detection in Videos (CVPRW)
2026๋
3์ 29์ผ
๐ Docs
[ Prometheus Project ] ๋ฅํ์ดํฌ ํ์ง
2026๋
3์ 28์ผ
๐ Study
Flash Attention: CPU-GPU์ BottleNeck
2026๋
3์ 25์ผ
๐ Study
![[ MCATrack ] Transformation Matrix ๊ณผ Feature Extraction](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ad69db3d0-f510-4f90-9969-2ad60a24a4ab%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-803a-a372-eddcc30369c5%26cache%3Dv2&w=3840&q=75)
[ MCATrack ] Transformation Matrix ๊ณผ Feature Extraction
2026๋
3์ 24์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] MCATrack : Tracking Tiny Drones against Clutter: Large-Scale Infrared Benchmark with Motion-Centric Adaptive Algorithm](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ab833606f-d86a-45ab-ba22-92e372c0bff2%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80dc-9f67-eca565337d49%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] MCATrack : Tracking Tiny Drones against Clutter: Large-Scale Infrared Benchmark with Motion-Centric Adaptive Algorithm
2026๋
3์ 23์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] MemLoTrack : Enhancing TIR Anti-UAV Tracking with Memory-Integrated Low-Rank Adaptation](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A0d7e47a7-dd70-4466-8e06-ec7d1e4a0bb0%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-808c-b03a-fdf776f88821%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] MemLoTrack : Enhancing TIR Anti-UAV Tracking with Memory-Integrated Low-Rank Adaptation
2026๋
3์ 9์ผ
๐ Paper Review
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Consistency Models
2026๋
2์ 17์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] A Simple Detector with Frame Dynamics is a Strong Tracker](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A517345e0-96ef-4aa1-ab08-10da5ffbaccc%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80a4-a7cf-f4ecf831b86d%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] A Simple Detector with Frame Dynamics is a Strong Tracker
2026๋
2์ 10์ผ
๐ Study
Dilation, linear-interpolation ๊ทธ๋ฆฌ๊ณ Bilinear upsampling
2026๋
2์ 9์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] ViT : An Image is Worth 16x16 Words : Transformers For Image Recognition At Scale](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A3022e39b-ac8f-435e-a2c9-ae0756eed609%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-8038-bbef-f5bd2af84500%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] ViT : An Image is Worth 16x16 Words : Transformers For Image Recognition At Scale
2026๋
2์ 7์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] DnCNN : Denoising Convolutional Neural Network](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253A48202177-e65c-4bc2-b6d9-64d01876aa62%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-808d-a551-ca8b78d21008%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] DnCNN : Denoising Convolutional Neural Network
2026๋
1์ 13์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] Word2Vec](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ab5016c23-ac7b-460d-9c38-2b3217162ecb%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80f0-a429-f13833da9e30%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Word2Vec
2025๋
12์ 14์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ] Attention Is All You Need](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Aa8d91cab-bf3c-41f8-846e-1357192fc31e%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80f5-912b-e35fef941f4c%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ] Attention Is All You Need
2025๋
11์ 9์ผ
๐ Paper Review
![[๋
ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] AlexNet with ImageNet](/_next/image?url=https%3A%2F%2Fwww.notion.so%2Fimage%2Fattachment%253Ad29a125b-9b23-4496-96ce-54c1a3de4d25%253Aimage.png%3Ftable%3Dblock%26id%3D391623bd-cb86-80a8-898e-c1918266c988%26cache%3Dv2&w=3840&q=75)
[๋ ผ๋ฌธ ๋ฆฌ๋ทฐ + ์ฝ๋ ๊ตฌํ] AlexNet with ImageNet
2025๋
9์ 1์ผ
์ค๋์ AlexNet์ ๋ ผ๋ฌธ ๋ฆฌ๋ทฐ์ ๋๋ค.
๐ Study
SIFT ํน์ง์ ๊ฒ์ถ
2025๋
8์ 6์ผ
SIFT๋ฅผ ํ๋ง๋๋ก ํํํ์๋ฉด ๋ค์๊ณผ ๊ฐ๋ค. ํฌ๊ธฐ, ํ์ , ์กฐ๋, affine์ ๋ณํ ๋ฐ noise์ ๋ถ๋ณํ๋ ํน์ง์ ์ถ์ถํ๋ ์๊ณ ๋ฆฌ์ฆ. ์ด ํฌ์คํธ๋ SIFT์ ๋ํ ๋ด์ฉ์ ๋๋ค.
๐ Docs
Welcome to morethan log!
2023๋
1์ 22์ผ