We propose product quantization adversarial generation (PQAG), a simple yet effective method to generate adversarial examples for product quantization based retrieval systems.

We obtain a complete characterization for Singlet-optimal LRCs with r = 2 and d = 6.

We construct a new class of optimal q-ary cyclic r-local LRCs with lengths n | q + 1 and a new class of optimal q-ary cyclic (r, δ)-LRCs (δ ≥ 2) with lengths n | q + 1.

We give an alternate proof of the Singleton-like bound for LRCs first proved by Gopalan et al. Some structural properties on optimal LRCs that achieve the Singleton-like bound are given.

We develop an adaptive stopping strategy that adapts our method to diverse images.

We propose universal adversarial head (UAH), which crafts adversarial query videos by prepending the original videos with a sequence of adversarial frames to perturb the normal hash codes in the Hamming space.

we have completely obtained all optimal (r, δ)-LRCs with length n | (q + 1) and (r + δ - 1) | n for all possible parameters for the completeness in the coding theory.

We establish important connections of the existence of LRCs with finite geometry and finite fields, and two systematic constructions of perfect LRCs are obtained.

We develop a novel method to solve the unbalanced variance problem combined with PQ and optimized product quantization (OPQ), called Mean-Removed Product Quantization (MRPQ) for the actual scenario.

Previous Model Inversion attacks have solely disclosed private information in the latent space of GAN priors, limiting their semantic extraction and transferability. Thus we propose Intermediate Features enhanced Generative Model Inversion (IF-GMI), which disassembles the GAN structure and exploits features between intermediate blocks.

The generative adversarial attacks use class labels as conditions, failing to leverage the semantic information. Thus, we propose CLIP-guided Generative Network with Cross-attention (CGNC) to enhance multi-target attacks by using textual knowledge of CLIP.

To solve the efficiency problem of Partially Relevant Video Retrieval (PRVR) methods, we proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly.

To learn compact 3D representations, we propose a simple yet effective Point Feature Enhancement Masked Autoencoders (Point-FEMAE), which mainly consists of a global branch and a local branch to capture latent semantic features.

To extract universal 3D vision-language embedding, we propose a vision-language pre-training framework 3DVLP (3D vision-language pre-training with object contrastive learning), which transfers flexibly on 3D vision-language downstream tasks.

We attempt to adapt the recognizer to low-resolution inputs by transferring the knowledge from the high-resolution, and propose an efficient and effective knowledge distillation framework to achieve multi-level knowledge transfer.

To improve the gradient inversion attack for Federated Learning by GAN model, we propose Gradient Inversion over Feature Domains(GIFD), which disassembles the GAN model and searches the feature domains of the intermediate layers.

We propose an Instance-aware Dynamic Prompt Tuning (IDPT) for point cloud pre-trained models, which utilizes a prompt module to perceive the semantic prior features of each instance.

We propose MultiScale Feature Domain Patch Matching (MSFDPM) to fully utilizes side information at the decoder of the distributed image compression model.

We propose a simple yet effective onestage SSVH method called ConMH, which incorporates video semantic information and video similarity relationship understanding in a single stage.

We introduce a new pretext task, Turning to Video for Transcript Sorting (TVTS), which sorts shuffled ASR scripts by attending to learned video representations.

We propose a general frequency-oriented framework (FSR) to accelerate SR networks by considering data characteristics in frequency domain.

We propose an adaptive gradient compression strategy called DAGC, which assigns each worker a different compression ratio according to their data volumes.

We propose Compressed Sensing based Asymmetric Semantic Image Compression (CS-ASIC) for resource-constrained IoT systems, which consists of a lightweight front encoder and a deep iterative decoder offloaded at the server.

We learn unsupervised binary descriptors by contrastive learning, which can better capture discriminative visual semantics.

We propose the first quantized representation learning method for cross-view video retrieval, namely Hybrid Contrastive Quantization (HCQ).

We propose Weakly-Supervised Deep Hyperspherical Quantization (WSDHQ), which is the first work to learn deep quantization from weakly tagged images.

Most existing CNN-based methods neglect the diversity of image contents and degradations in the corrupted images. We propose deep mix-order attention networks (MAN) to extract features that capture rich feature statistics within networks.

We propose a novel method, dubbed deep hashing targeted attack (DHTA), to study the targeted attack on the deep hashing based retrieval.

We propose product quantization adversarial generation (PQ-AG), a simple yet effective method to generate adversarial examples for product quantization based retrieval systems.

We propose an effective Deep Image Prior Driven Defense (DIPDefend) method against adversarial examples.