Juneta TaoOne-2–3–45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape OptimisationSingle image 3D reconstruction is crucial for a wide range of applications, such as robotic object manipulation and navigation, 3D content…·3 min read·Jun 12, 2024----
Juneta TaoImage Based Anomaly DetectionImage Based Anomaly Detection (IAD) is an important computer vision task for industrial manufacturing (IM) applications , such as…·5 min read·Jun 6, 2024----
Juneta TaoInteresting Workshops And Tutorials in CVPR 2024CVPR 2024 will take place in Seattle on 16th to 21st June. To prepare for attending the conference. Here listed some interesting workshops…·3 min read·May 28, 2024----
Juneta TaoImage Based Virtual Try-On (VITON)The goal of image based VITON is to fit a given cloth image to a person in a reference image. The majority methods include two steps. First…·5 min read·Apr 23, 2024--2--2
Juneta TaoAdapt Segment Anything for Medical Image SegmentationSegment Anything Model (SAM) shows promising zero-shot image segmentation performance on natual images. There are also quite a few papers…·3 min read·Apr 5, 2024----
Juneta TaoLabelling Dataset with Ground SAMGround SAM [1] is proposed to combine multiple foundation models to perform object detection, segmentation, and in-painting with text…·3 min read·Mar 25, 2024----
Juneta TaoParameter-efficient Finetuning — Visual Prompt TuningVisual Prompt Tuning (VPT) [1] is one of the efficient ways to adapt large pre-trained Transformers to downstream tasks in terms of…·4 min read·Mar 19, 2024----
Juneta TaoCreate/Expand Dataset with Generative ModelsLarge scale diverse dataset is one of the key element to train a robust model. Collecting and labelling real-world datasets is time…·5 min read·Mar 4, 2024----
Juneta TaoYOLOv9: Learning What You Want to Learn Using Programmable Gradient InformationYOLOv9 proposed two modules, Programmable Gradient Information (PGI) and Generalised ELAN (GELAN), to boost the performance for real time…·3 min read·Feb 27, 2024----
Juneta TaoDepth Anything: Unleashing the Power of Large-Scale Unlabeled DataFoundation models demonstrated strong performance on zero-/fewshot scenarios. To build foundation model for Monocular Depth Estimation…·4 min read·Feb 27, 2024----