視覺模式分析及實踐
李平
- 出版商: 電子工業
- 出版日期: 2026-06-01
- 定價: $599
- 售價: $598
- 語言: 簡體中文
- 頁數: 400
- ISBN: 712152936X
- ISBN-13: 9787121529368
-
相關分類:
Computer Vision
下單後立即進貨 (約4週~6週)
商品描述
本書聚焦視覺模式分析領域，以視覺數據為研究對象，屬於計算機視覺和模式識別的範疇。全書系統闡述視覺模式分析的基礎模型和典型任務的算法原理，並通過 PyTorch 代碼實例講解算法的核心思想，助力讀者理解與實踐。全書共分10章：第1章簡要介紹視覺模式分析領域的發展歷程和前沿進展，為讀者構建領域整體認知。第2章詳細介紹主流的深度學習骨幹網絡，包括 CNN、RNN（GRU、LSTM、Mamba）、GNN、GAN、Transformer、Diffusion model 等，夯實理論基礎。第3章至第10章分別圍繞視頻動作識別、動作檢測（時序和時空角度）、視頻描述、視頻定位、視覺語義分割、視覺模型壓縮、視覺對抗攻防等典型任務展開，每個任務均介紹重要算法和代碼實例，涵蓋全監督、半監督、弱監督等多種場景，以及模型壓縮和對抗攻防等前沿方向。
目錄大綱
第 1 章 緒論·································································································1
1.1 視頻動作理解 ·····················································································2
1.2 視頻描述與定位··················································································2
1.3 視覺語義分割 ·····················································································3
1.4 視覺模型輕量化··················································································3
1.5 視覺模型安全性··················································································4
1.6 本章習題 ···························································································4
參考文獻 ································································································6
第 2 章 基礎網絡結構 ·····················································································8
2.1 CNN ································································································8
2.1.1 AlexNet ·························································································8
2.1.2 ResNet ··························································································9
2.2 VGG ·····························································································.11
2.3 RNN ·····························································································.12
2.3.1 LSTM ························································································.12
2.3.2 GRU ··························································································.13
2.4 Diffusion ························································································.15
2.5 Transformer ·····················································································.16
2.6 GNN ·····························································································.17
2.7 Mamba ··························································································.18
2.8 GAN ·····························································································.19
2.9 本章習題 ························································································.19
參考文獻 ·····························································································.20
第 3 章 動作識別 ························································································.21
3.1 任務介紹 ························································································.21
VI | 視覺模式分析及實踐
3.2 方法總覽 ························································································.22
3.2.1 概況 ··························································································.22
3.2.2 基於雙流卷積神經網絡的動作識別 ···················································.22
3.2.3 基於三維卷積神經網絡的方法 ·························································.24
3.2.4 基於循環神經網絡的動作識別 ·························································.24
3.2.5 基於 Transformer 的方法 ·································································.25
3.2.6 全監督點雲動作識別 ·····································································.25
3.2.7 半監督點雲動作識別 ·····································································.26
3.2.8 自監督點雲動作識別 ·····································································.27
3.3 典型方法 ························································································.27
3.3.1 基於雙流卷積神經網絡的 RGB 動作識別 ···········································.27
3.3.2 基於三維卷積神經網絡的 RGB 動作識別 ···········································.32
3.3.3 基於循環神經網絡的 RGB 動作識別 ·················································.34
3.3.4 基於 Transformer 的 RGB 動作識別 ··················································.35
3.3.5 全監督點雲動作識別 ·····································································.39
3.3.6 半監督點雲動作識別 ····································································.42
3.3.7 自監督點雲動作識別 ·····································································.44
3.4 代碼實例 ·······················································································.47
3.4.1 RGB 動作識別方法 ·······································································.47
3.4.2 基於膨脹三維卷積神經網絡的動作識別 ·············································.52
3.4.3 基於點 4D Transformer 的點雲動作識別 ·············································.56
3.4.4 基於掩碼偽標記自編碼的半監督點雲動作識別 ····································.59
3.5 本章習題 ························································································.64
3.5.1 簡答題 ·······················································································.64
3.5.2 動手實踐題 ·················································································.64
參考文獻 ·····························································································.66
第 4 章 時序動作檢測 ··················································································.69
4.1 任務介紹 . ······················································································.69
4.2 方法總覽 ························································································.70
4.2.1 概況 ··························································································.70
4.2.2 全監督時序動作檢測 ·····································································.70
4.2.3 半監督時序動作檢測 ·····································································.74
4.2.4 點監督時序動作檢測 ·····································································.75
4.2.5 弱監督時序動作檢測 ·····································································.75
4.3 典型方法 ························································································.76
4.3.1 全監督時序動作檢測 ·····································································.76
4.3.2 半監督時序動作檢測 ·····································································.80
4.3.3 點監督時序動作檢測 ·····································································.83
4.3.4 弱監督時序動作檢測 ·····································································.88
4.4 代碼實例 ························································································.94
4.4.1 全監督時序動作檢測 ·····································································.94
4.4.2 半監督時序動作檢測 ·····································································100
4.4.3 點監督時序動作檢測 ·····································································104
4.4.4 弱監督時序動作檢測 ·····································································109
4.5 本章習題 ························································································111
4.5.1 簡答題 ······················································································.111
4.5.2 動手實踐題 ················································································.111
參考文獻 ·····························································································113
第 5 章 時空動作檢測 ··················································································115
5.1 任務介紹 ························································································115
5.2 方法總覽 ························································································116
5.2.1 概況 ··························································································116
5.2.2 全監督時空動作檢測 ·····································································116
5.2.3 半監督時空動作檢測 ·····································································119
5.2.4 點監督時空動作檢測 ·····································································120
5.2.5 弱監督時空動作檢測 ·····································································120
5.3 典型方法 ························································································121
5.3.1 全監督時空動作檢測 ·····································································121
5.3.2 半監督時空動作檢測 ·····································································125
5.3.3 點監督時空動作檢測 ·····································································127
5.3.4 弱監督時空動作檢測 ·····································································131
5.4 代碼實例 ························································································135
5.4.1 全監督時空動作檢測 ·····································································135
5.4.2 半監督時空動作檢測 ·····································································139
5.5 本章習題 ························································································144
5.5.1 簡答題 ·······················································································144
VIII | 視覺模式分析及實踐
5.5.2 動手實踐題 ·················································································145
參考文獻 ·····························································································146
第 6 章 視頻描述 ························································································149
6.1 任務介紹 ························································································149
6.2 方法總覽 ························································································150
6.2.1 方法概述 ····················································································150
6.2.2 全監督單語句視頻描述 ··································································150
6.2.3 密集視頻描述 ··············································································152
6.2.4 少樣本及少監督視頻描述 ·······························································152
6.2.5 零樣本視頻描述 ···········································································153
6.3 典型方法 ························································································154
6.3.1 單語句視頻描述 ···········································································154
6.3.2 密集視頻描述 ··············································································156
6.3.3 少監督視頻描述 ···········································································158
6.4 代碼實例 ························································································162
6.4.1 單語句視頻描述 ···········································································162
6.4.2 零樣本視頻描述 ···········································································166
6.5 本章習題 ························································································170
6.5.1 簡答題 ·······················································································170
6.5.2 動手實踐題 ·················································································170
參考文獻 ·····························································································171
第 7 章 視頻定位 ························································································174
7.1 任務介紹 ························································································174
7.2 方法總覽 ························································································175
7.2.1 概況 ··························································································175
7.2.2 全監督視頻定位 ···········································································175
7.2.3 弱監督視頻定位 ···········································································178
7.2.4 點監督視頻定位 ···········································································179
7.3 典型方法 ························································································180
7.3.1 全監督視頻定位 ···········································································180
7.3.2 弱監督視頻定位 ···········································································185
7.3.3 點監督視頻定位 ···········································································191
7.4 代碼實例 ························································································198
7.4.1 全監督視頻定位 ···········································································198
7.4.2 弱監督視頻定位 ···········································································203
7.5 本章習題 ························································································209
7.5.1 簡答題 ·······················································································209
7.5.2 動手實踐題 ·················································································210
參考文獻 ·····························································································211
第 8 章 語義分割 ························································································214
8.1 任務介紹 ························································································214
8.2 方法總覽 ························································································215
8.2.1 概況 ··························································································215
8.2.2 全監督圖像語義分割 ·····································································216
8.2.3 半監督圖像語義分割 ·····································································217
8.2.4 弱監督圖像語義分割 ·····································································219
8.2.5 RGB-T 語義分割 ···········································································220
8.2.6 RGB-D 語義分割 ··········································································221
8.2.7 參考圖像語義分割 ········································································222
8.2.8 視頻語義分割 ··············································································225
8.3 典型方法 ························································································227
8.3.1 全監督圖像語義分割 ·····································································227
8.3.2 半監督圖像語義分割 ·····································································231
8.3.3 弱監督圖像語義分割 ·····································································234
8.3.4 RGB-T 語義分割 ···········································································237
8.3.5 RGB-D 語義分割 ··········································································239
8.3.6 參考語句分割 ··············································································242
8.3.7 視頻語義分割 ··············································································245
8.4 代碼實例 ······················································································.248
8.4.1 全監督圖像語義分割 ·····································································248
8.4.2 半監督圖像語義分割 ·····································································251
8.4.3 RGB-D 語義分割 ··········································································254
8.4.4 參考語句分割 ··············································································261
8.4.5 視頻語義分割 ··············································································264
8.5 本章習題 ························································································268
8.5.1 簡答題 ·······················································································268
8.5.2 動手實踐題 ·················································································268
參考文獻 ·····························································································270
第 9 章 模型壓縮 ························································································275
9.1 任務介紹 ························································································275
9.2 方法總覽 ························································································276
9.2.1 方法概述 ····················································································276
9.2.2 知識蒸餾 ····················································································277
9.2.3 網絡剪枝 ····················································································281
9.2.4 網絡量化 ····················································································284
9.2.5 數據集蒸餾 ·················································································287
9.3 典型方法 ························································································290
9.3.1 知識蒸餾 ····················································································290
9.3.2 網絡剪枝 ····················································································294
9.3.3 網絡量化 ····················································································296
9.3.4 數據集蒸餾 ·················································································300
9.4 代碼實例 ························································································303
9.4.1 基於響應或中間特征的知識蒸餾方法 ················································303
9.4.2 基於跨頭知識蒸餾的目標檢測 ·························································304
9.4.3 網絡剪枝 ····················································································306
9.4.4 網絡量化 ····················································································309
9.4.5 數據集蒸餾 ·················································································313
9.5 本章習題 ························································································319
9.5.1 簡答題 ·······················································································319
9.5.2 動手實踐題 ·················································································320
參考文獻 ·····························································································323
第 10 章 對抗攻防 ······················································································327
10.1 任務介紹 ······················································································327
10.2 方法總覽 ······················································································328
10.2.1 視覺白盒攻擊 ·············································································329
10.2.2 視覺黑盒查詢攻擊 ·······································································330
10.2.3 視覺黑盒遷移攻擊 ·······································································331
10.2.4 視覺對抗訓練防禦方法 ·································································334
10.2.5 視覺對抗凈化防禦方法 ·································································337
10.2.6 大模型攻擊方法 ··········································································338
10.2.7 大模型防禦方法 ··········································································343
10.3 典型方法 ······················································································345
10.3.1 白盒攻擊 ···················································································345
10.3.2 查詢攻擊 ···················································································347
10.3.3 遷移攻擊 ···················································································348
10.3.4 視覺對抗訓練防禦方法 ·································································349
10.3.5 視覺對抗凈化防禦方法 ·································································353
10.3.6 大模型攻擊方法 ··········································································354
10.3.7 大模型防禦方法 ··········································································357
10.4 代碼實例 ······················································································360
10.4.1 白盒攻擊 ···················································································360
10.4.2 查詢攻擊 ···················································································361
10.4.3 遷移攻擊 ···················································································362
10.4.4 視覺對抗訓練防禦方法 ·································································364
10.4.5 視覺對抗凈化防禦方法 ·································································368
10.4.6 大模型攻擊方法 ··········································································371
10.4.7 大模型防禦方法 ··········································································374
10.5 本章習題 ······················································································377
10.5.1 簡答題 ······················································································377
10.5.2 動手實踐題 ················································································378
參考文獻 ·····························································································380
