The target of video moment retrieval (VMR) is predicting temporal spans within a video that semantically match a given linguistic query. Existing VMR methods based on multimodal large language models (MLLMs) overly rely on expensive high-quality datasets and time-consuming fine-tuning. Although some recent studies introduce a zero-shot setting to avoid fine-tuning, they overlook inherent language bias in the query, leading to erroneous localization. To tackle the aforementioned challenges, this paper proposes Moment-GPT, a tuning-free pipeline for zero-shot VMR utilizing frozen MLLMs. Specifically, we first employ LLaMA-3 to correct and rephrase the query to mitigate language bias. Subsequently, we design a span generator combined with MiniGPT-v2 to produce candidate spans adaptively. Finally, to leverage the video comprehension capabilities of MLLMs, we apply Video-ChatGPT and span scorer to select the most appropriate spans. Our proposed method substantially outperforms the state-of-the-art MLLM-based and zero-shot models on several public datasets, including QVHighlights, ActivityNet-Captions, and Charades-STA.

Recent advancements in diffusion-based technologies have made significant strides, particularly in identity-preserved portrait generation (IPG). However, when using multiple reference images from the same ID, existing methods typically produce lower-fidelity portraits and struggle to customize face attributes precisely. To address these issues, this paper presents HiFi-Portrait, a high-fidelity method for zero-shot portrait generation. Specifically, we first introduce the face refiner and landmark generator to obtain fine-grained multi-face features and 3D-aware face landmarks. The landmarks include the reference ID and the target attributes. Then, we design HiFi-Net to fuse multi-face features and align them with landmarks, which improves ID fidelity and face control. In addition, we devise an automated pipeline to construct an ID-based dataset for training HiFi-Portrait. Extensive experimental results demonstrate that our method surpasses the SOTA approaches in face similarity and controllability. Furthermore, our method is also compatible with previous SDXL-based works.

Given a video and a linguistic query, video moment retrieval and highlight detection (MR&HD) aim to locate all the relevant spans, while simultaneously predicting saliency scores. Most existing methods utilize RGB images as input, overlooking the inherent multi-modal visual signals like optical flow and depth. In this paper, we propose a Multi-modal Fusion and Query Refinement Network (MRNet) to learn complementary information from multi-modal cues. Specifically, we design a multi-modal fusion module to dynamically combine RGB, optical flow, and depth map. Furthermore, to simulate human understanding of sentences, we introduce a query refinement module that merges text at different granularities, containing word-, phrase-, and sentence-wise levels. Comprehensive experiments on QVHighlights and Charades datasets indicate that MRNet outperforms current SOTA methods, achieving notable improvements in MR-mAP@Avg (+3.41) and HD-HIT@1 (+3.46) on QVHighlights.

With the increasing demand for video understanding, video moment and highlight detection (MHD) has emerged as a critical research topic. MHD aims to localize all moments and predict clip-wise saliency scores simultaneously. Despite progress made by existing DETR-based methods, we observe that these methods coarsely fuse features from different modalities, which weakens the temporal intra-modal context and results in insufficient cross-modal interaction. To address this issue, we propose MH-DETR (Moment and Highlight DEtection TRansformer) tailored for MHD. Specifically, we introduce a simple yet efficient pooling operator within the uni-modal encoder to capture global intra-modal context. Moreover, to obtain temporally aligned cross-modal features, we design a plug-and-play cross-modal interaction module between the encoder and decoder, seamlessly integrating visual and textual features. Comprehensive experiments on QVHighlights, Charades-STA, Activity-Net, and TVSum datasets show that MH-DETR outperforms existing state-of-the-art methods, demonstrating its effectiveness and superiority. Our code is available at https://github.com/YoucanBaby/MH-DETR.

Moment retrieval (MR) and highlight detection (HD) aim to identify relevant moments and highlights in video from corresponding natural language query. Large language models (LLMs) have demonstrated proficiency in various computer vision tasks. However, existing methods for MR&HD have not yet been integrated with LLMs. In this letter, we propose a novel two-stage model that takes the output of LLMs as the input to the second-stage transformer encoder-decoder. First, MiniGPT-4 is employed to generate the detailed description of the video frame and rewrite the query statement, fed into the encoder as new features. Then, semantic similarity is computed between the generated description and the rewritten queries. Finally, continuous high-similarity video frames are converted into span anchors, serving as prior position information for the decoder. Experiments demonstrate that our approach achieves a state-of-the-art result, and by using only span anchors and similarity scores as outputs, positioning accuracy outperforms traditional methods, like Moment-DETR.

With the surge in online video content, finding highlights and key video segments have garnered widespread attention. Given a textual query, video highlight detection (HD) and temporal grounding (TG) aim to predict frame-wise saliency scores from a video while concurrently locating all relevant spans. Despite recent progress in DETR-based works, these methods crudely fuse different inputs in the encoder, which limits effective cross-modal interaction. To solve this challenge, the authors design QD-Net (query-guided refinement and dynamic spans network) tailored for HD&TG. Specifically, they propose a query-guided refinement module to decouple the feature encoding from the interaction process. Furthermore, they present a dynamic span decoder that leverages learnable 2D spans as decoder queries, which accelerates training convergence for TG. On QVHighlights dataset, the proposed QD-Net achieves 61.87 HD-HIT@1 and 61.88 [email protected], yielding a significant improvement of +1.88 and +8.05, respectively, compared to the state-of-the-art method.