Jul 2026· International Symposium on Communications and Information Technologies· pp. 1-5· 0 citations· 13 references
Computer Science
TL;DR
A method combining deformable temporal alignment and difference-aware spatial selective fusion is proposed, used to generate complementary temporal context and achieves certain rate-distortion performance improvement over DCVC-DC.
Abstract
In conditional coding-based neural video compression, the quality of temporal context directly affects compression performance. Existing methods mostly construct context from propagated reference features, but they are vulnerable to motion estimation and local alignment errors in regions with complex motion, occlusion, and high-frequency textures, resulting in inaccurate temporal information. To address this issue, this paper proposes a method combining deformable temporal alignment and difference-aware spatial selective fusion. A Context-aware Temporal Alignment Module is used to generate complementary temporal context, while a Difference-aware Spatial Selective Fusion module adaptively selects reliable temporal information and suppresses misalignment. Experiments show that the proposed method achieves certain rate-distortion performance improvement over DCVC-DC.
High dynamic range (HDR) video reconstruction from al\-ter\-na\-ting-exposure sequences remains challenging, especially in regions with extreme luminance variation. We propose a novel HDR reconstruction framework based on dual-stream registration and accurate pyramid fusion. Given three consecutive frames, our method c...
O. Martorell, Ivan Pereira-Sánchez, Antoni Fuentes et al.· 0 citations
Different from natural videos, Screen Content Videos (SCVs) are characterized by abrupt motion, scene switches, and high-frequency details such as text and graphics. Conventional video enhancement methods, which rely heavily on temporal continuity, often suffer from performance degradation when processing SCVs due to t...
Zi-Yin Huang, Sik-Ho Tsang, Xin Qin et al.· 0 citations
Neural video codecs derive motion and temporal contexts mainly from RGB frames, leaving room for cross-modal guidance from complementary temporal observations. Event streams can provide such observations by recording brightness changes between frames. In this work, we propose an Event-guided Neural Video Codec (ENVC) t...
Ji-Yun Kong, Jungwoo Kim, Enes Eray Demirtas et al.· 0 citations
The modern tools used for video editing and compression make video tampering detection extremely difficult for digital forensic investigators, as it can often hide visual and temporal evidence or artifacts of tampering. In this work, we propose TamperNet, a hybrid spatio-temporal framework to detect various video man...
M. Mujtaba, Muhammad Imran, S. Irfan et al.· Neural Processing Letters· 0 citations
Spatio-Temporal Video Super-Resolution (STVSR) reconstructs high-resolution frames in space and time from low-resolution inputs. Existing models often degrade when motion dynamics at test time differ from training data. We propose a Test-Time Adaptation (TTA) framework that fine-tunes a pretrained STVSR decoder using o...
The use of artificial intelligence and smart algorithms for the restoration, enhancement and creation of videos from historical images continues to grow. To tackle the problems of spatial structure drift, inter-frame flickering, and motion discontinuity in the animation of historical images, we propose a video diffusio...
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.
