T2V-Turbo breaks video consistency model quality bottleneck with mixed reward feedback, achieving superior 4-step generation over 50-step DDIM.

Jiachen Li, Weixi Feng, Tsu-Jui Fu et al.

EffoVPR leverages DINOv2 self-attention features for zero-shot and fine-tuned VPR, achieving state-of-the-art results with compact 128D features.

Issar Tzachor, Boaz Lerner, Matan Levy et al.

MoSca: a 4D dynamic scene reconstruction system using Gaussian fusion from casual monocular videos, leveraging a novel motion scaffold and physics-inspired regularization, achieving state-of-the-art results.

Jiahui Lei, Yijia Weng, Adam Harley et al.

MambaLLIE uses global-then-local state space design to achieve PSNR of 30.12 in low-light image enhancement.

Jiangwei Weng, Zhiqiang Yan, Ying Tai et al.

LucidPPN enhances user understanding in computer vision by separating color from other visual features.

Mateusz Pach, Dawid Rymarczyk, Koryna Lewandowska et al.

FIFO-Diffusion leverages pretrained diffusion models with diagonal denoising, latent partitioning, and lookahead to generate infinite long videos without extra training.

Jihwan Kim, Junoh Kang, Jinyoung Choi et al.

Integrating point clouds and NeRF with LLMs enhances 3D scene understanding and reasoning.

Xianzheng Ma, Brandon Smart, Yash Bhalgat et al.

BoQ uses learnable global queries with cross-attention, outperforming SOTA in visual place recognition with high speed and efficiency.

Amar Ali-Bey, Brahim Chaib-draa, Philippe Giguère

This survey organizes video representation learning around RGB, optical flow, CNNs, RNNs, and GCNs, without reporting unified benchmark scores.

Elham Ravanbakhsh, Yongqing Liang, J. Ramanujam et al.

Diff-IP2D employs diffusion models for joint 2D hand trajectory and object affordance prediction, outperforming SOTA with 10% FDE improvement.

Junyi Ma, Jingyi Xu, Xieyuanli Chen et al.

MANTIS enhances multi-image vision language skills via instruction tuning, achieving 13-point improvement.

Dongfu Jiang, Xuan He, Huaye Zeng et al.

EchoScene employs a dual-branch diffusion model with information echo to generate 3D indoor scenes from scene graphs, enhancing controllability and global coherence.

Guangyao Zhai, Evin Pınar Örnek, Dave Zhenyu Chen et al.

DOCCI dataset with 15k detailed image descriptions enhances fine-grained vision-language tasks.

Yasumasa Onoe, Sunayana Rane, Zachary Berger et al.

GS-LRM is a Transformer-based large-scale 3D reconstruction model that predicts high-quality scene primitives from 2-4 sparse views in 0.23 seconds.

Kai Zhang, Sai Bi, Hao Tan et al.

Proposed a multi-page Document VQA method using self-attention scoring, achieving state-of-the-art performance without OCR.

Lei Kang, Rubèn Tito, Ernest Valveny et al.

LM-IGTD transforms low-dimensional and mixed-type tabular data into images using noise augmentation and improved IGTD, boosting CNN classification performance.

Vanesa Gómez-Martínez, Francisco J. Lara-Abelenda, Pablo Peiro-Corbacho et al.

Proposes parameter-free pooling in PLLaVA to extend image pretraining to dense video captioning, achieving SOTA results.

Lin Xu, Yilin Zhao, Daquan Zhou et al.

ViViDex leverages human videos, combining trajectory-guided reinforcement learning and visual distillation, to enable dexterous multi-finger robot manipulation with minimal demonstration data.

Zerui Chen, Shizhe Chen, Etienne Arlaud et al.

Proposes 'Align Your Steps' framework to optimize diffusion model sampling schedules, significantly improving quality in few-step generation.

Amirmojtaba Sabour, Sanja Fidler, Karsten Kreis

Retinexformer combines Transformer-based multi-scale feature fusion and illumination-guided modules, achieving PSNR of 25.30dB on ultra-high-resolution low-light images.

Xiaoning Liu, Zongwei Wu, Ao Li et al.