Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks34 from public data
- SoccerNet 2022 Challenges Results42
The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team, and tasks (3-6) were novel, including their underlying data and annotations.
- LLM-Detector: Improving AI-Generated Chinese Text Detection with Open-Source LLM Instruction Tuning31
LLM-Detector is proposed, a novel method for both document-level and sentence-level text detection through Instruction Tuning of LLMs that not only significantly outperforms baseline methods in both sentence-level and document-level text detection but also demonstrates strong generalization capabilities.
- 29
- Contextual Multi-Scale Feature Learning for Person Re-Identification21
CMSNet is the most lightweight model for person Re- ID and it achieves state-of-the-art performance on four commonly used Re-ID datasets, surpassing most large-scale models.
- AI-VQA14
This paper proposes Agent Interaction Visual Question Answering (AI-VQA), a task investigating deep scene understanding if the agent takes a certain action, and proposes a novel method, called ARE, that can comprehend the interaction and explain the reason based on a given event knowledge base.
- Dropletvideo: A Dataset and Approach to Explore Integral Spatio-Temporal Consistent Video Generation10
This paper introduces and examines integral spatio-temporal consistency, considering the synergy between plot progression and camera techniques, and the long-term impact of prior content on subsequent generation.
- 8
- Strong generalization in quantum neural networks8
It is proved that Qnns have a generalization error bound and proposed its theoretical value, and it is shown thatQnns perform almost the same on the training dataset and test dataset without the overfitting phenomenon.
- Knowledge-Supervised Learning: Knowledge Consensus Constraints for Person Re-Identification8
A novel Knowledge-Supervised Learning (KSL) method for person re-identification (Re-ID), which can improve the performance without introducing extra inference cost, and is introduced isomorphic auxiliary training strategy to conduct basic multiple views that simultaneously train multiple classifier heads of the same network on the same training data.
- 8
- 7
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 3
- An OpenCLTM Implementation of WebP Accelerator on FPGAs3
An OpenCL implementation of WebP accelerator on FPGAs to optimize the performance of Webp Lossy Compression Algorithm and is more competitive for data centers to achieve higher performance and lower cost.
- Droplet3D: Commonsense Priors from Videos Facilitate 3D Generation2
This paper introduces Droplet3D-4M, the first large-scale video dataset with multi-view level annotations, and introduces Droplet3D, a generative model supporting both image and dense text input, which exhibits the potential for extension to scene-level applications.
- MVIndEmo: a dataset for micro video public-induced emotion prediction on social media2
This work has put forth two novel tasks concerning the recognition of public-induced emotion on micro videos: emotion polarity and emotion classification, and introduced a accessible dataset specifically tailored for the analysis of public-induced emotion on micro videos.
- 2
- Towards Further Comprehension on Referring Expression with Rationale2
This work works on how to enable an artificial agent to understand RE further and proposes a more comprehensive task, called Further Comprehension on Referring Expression (FREC), which mainly focuses on three sub-tasks: correcting the erroneous text expression based on visual information, generating the rationale of this input expression and localizing the proper object based on the corrected expression.
- Pose-Aware Diffusion for 3D Generation1
P Pose-Aware Diffusion is introduced, a novel end-to-end diffusion framework that synthesizes 3D geometry directly within the observation space that achieves superior geometric alignment and image-to-3D correspondence compared to state-of-the-art methods.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.