https://scholar.dgist.ac.kr/handle/20.500.11750/12135 2026-07-06T19:13:33Z https://scholar.dgist.ac.kr/handle/20.500.11750/60432 Title: 문맥 인지 비디오 인스턴스 세그먼테이션 방법 Author(s): 이승훈; 최민우; 서지완; 한길준; 임성훈 https://scholar.dgist.ac.kr/handle/20.500.11750/60427 Title: 단안 카메라 이미지에 대한 깊이 추정 방법 Author(s): 최원혁; 임성훈; 신민규 Abstract: 본 개시의 일 실시예의 컴퓨터 판독가능 저장 매체에 저장된 컴퓨터 프로그램이 개시된다. 상기 컴퓨터 프로그램은, 하나 이상의 프로세서에서 실행되는 경우 객체의 깊이 추정을 위한 이하의 방법들을 수행하도록 하며, 상기 방법은, 피처 추출 네트워크에서 학습용 이미지 데이터에 포함된 둘 이상의 객체들 각각을 피처 공간에 피처 점(feature point)으로 매핑하는 단계; 상기 피처 공간에 매핑 된 각각의 피처 점들 중 적어도 일부의 피처 점들 사이의 피처 공간에서의 거리와 상기 둘 이상의 객체들 중 상기 피처 점들과 대응되는 적어도 일부의 객체들 사이의 깊이 공간에서의 거리를 비교하는 단계; 및 상기 비교 결과에 기초하여, 상기 피처 점들 사이의 피처 공간에서의 거리와 상기 피처 점들과 대응되는 적어도 일부의 객체들 사이의 깊이 공간에서의 거리가 연관되도록 상기 피처 추출 네트워크를 학습시키는 단계를 포함할 수 있다. https://scholar.dgist.ac.kr/handle/20.500.11750/60002 Title: Scale-Invariant and View-Relational Representation Learning for Full Surround Monocular Depth Author(s): Hwang, Kyumin; Choi, Wonhyeok; Han, Kiljoon; Choi, Wonjoon; Choi, Minwoo; Na, Yongcheon; Park, Minwoo; Im, Sunghoon Abstract: Recent foundation models demonstrate strong generalization capabilities in monocular depth estimation. However, directly applying these models to Full Surround Monocular Depth Estimation (FSMDE) presents two major challenges: (1) high computational cost, which limits realtime performance, and (2) difficulty in estimating metricscale depth, as these models are typically trained to predict only relative depth. To address these limitations, we propose a novel knowledge distillation strategy that transfers robust depth knowledge from a foundation model to a lightweight FSMDE network. Our approach leverages a hybrid regression framework combining the knowledge distillation scheme–traditionally used in classification–with a depth binning module to enhance scale consistency. Specifically, we introduce a crossinteraction knowledge distillation scheme that distills the scaleinvariant depth bin probabilities of a foundation model into the student network while guiding it to infer metric-scale depth bin centers from ground-truth depth. Furthermore, we propose view-relational knowledge distillation, which encodes structural relationships among adjacent camera views and transfers them to enhance cross-view depth consistency. Experiments on DDAD and nuScenes demonstrate the effectiveness of our method compared to conventional supervised methods and existing knowledge distillation approaches. Moreover, our method achieves a favorable trade-off between performance and efficiency, meeting real-time requirements. © 2016 IEEE. 2025-12-31T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/59362 Title: Towards Lossless Implicit Neural Representation via Bit Plane Decomposition Author(s): Han, Woo Kyoung; Lee, Byeonghun; Cho, Hyunmin; Im, Sunghoon; Jin, Kyong Hwan Abstract: We quantify the upper bound on the size of the implicit neural representation (INR) model from a digital perspective. The upper bound of the model size increases exponentially as the required bit-precision increases. To this end, we present a bit-plane decomposition method that makes INR predict bit-planes, producing the same effect as reducing the upper bound of the model size. We validate our hypothesis that reducing the upper bound leads to faster convergence with constant model size. Our method achieves lossless representation in 2D image and audio fitting, even for high bit-depth signals, such as 16-bit, which was previously unachievable. We pioneered the presence of bit bias, which INR prioritizes as the most significant bit (MSB). We expand the application of the INR task to bit depth expansion, lossless image compression, and extreme network quantization. Our source code is available at https: //github.com/WooKyoungHan/LosslessINR. 2025-06-12T15:00:00Z